wehub/anomalyco--models.dev
1
0
Fork 0
Code Issues Pull Requests Actions Deployments Models Agent Notes Packages Projects Releases Wiki Activity

chore: import upstream snapshot with attribution

Browse Source
This commit is contained in:
wehub-resource-sync
2026-07-13 12:28:55 +08:00
commit db42b91b75
6397 changed files with 146012 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
models/alibaba/qwen-flash.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Qwen Flash"
description = "Efficient Qwen model for fast chat, extraction, and high-volume workloads"
family = "qwen"
release_date = "2025-07-28"
last_updated = "2025-07-28"
attachment = false
reasoning = true
temperature = true
tool_call = true
knowledge = "2024-04"
open_weights = false
[limit]
context = 1_000_000
output = 32_768
[modalities]
input = ["text"]
output = ["text"]
models/alibaba/qwen-max.toml
+26
View File
@@ -0,0 +1,26 @@
name = "Qwen Max"
description = "Flagship Qwen model for complex reasoning, coding, and agentic workflows"
family = "qwen"
release_date = "2024-04-03"
last_updated = "2025-01-25"
attachment = false
reasoning = false
temperature = true
tool_call = true
knowledge = "2024-04"
open_weights = false
[limit]
context = 32_768
output = 8_192
[modalities]
input = ["text"]
output = ["text"]
[[benchmarks]]
name = "Aider Polyglot"
score = 21.8
metric = "percent correct"
source = "https://aider.chat/docs/leaderboards/"
date = "2025-01-28"
models/alibaba/qwen-omni-turbo.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Qwen-Omni Turbo"
description = "Qwen omni model for text, vision, audio, and multimodal agent tasks"
family = "qwen"
release_date = "2025-01-19"
last_updated = "2025-03-26"
attachment = false
reasoning = false
temperature = true
tool_call = true
knowledge = "2024-04"
open_weights = false
[limit]
context = 32_768
output = 2_048
[modalities]
input = ["text", "image", "audio", "video"]
output = ["text", "audio"]
models/alibaba/qwen-plus.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Qwen Plus"
description = "Qwen instruction model for multilingual chat, reasoning, and tool use"
family = "qwen"
release_date = "2024-01-25"
last_updated = "2025-09-11"
attachment = false
reasoning = true
temperature = true
tool_call = true
knowledge = "2024-04"
open_weights = false
[limit]
context = 1_000_000
output = 32_768
[modalities]
input = ["text"]
output = ["text"]
models/alibaba/qwen-turbo.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Qwen Turbo"
description = "Efficient Qwen model for fast chat, extraction, and high-volume workloads"
family = "qwen"
release_date = "2024-11-01"
last_updated = "2025-04-28"
attachment = false
reasoning = true
temperature = true
tool_call = true
knowledge = "2024-04"
open_weights = false
[limit]
context = 1_000_000
output = 16_384
[modalities]
input = ["text"]
output = ["text"]
models/alibaba/qwen-vl-max.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Qwen-VL Max"
description = "Qwen vision-language model for visual reasoning, documents, and agent tasks"
family = "qwen"
release_date = "2024-04-08"
last_updated = "2025-08-13"
attachment = false
reasoning = false
temperature = true
tool_call = true
knowledge = "2024-04"
open_weights = false
[limit]
context = 131_072
output = 8_192
[modalities]
input = ["text", "image"]
output = ["text"]
models/alibaba/qwen-vl-plus.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Qwen-VL Plus"
description = "Qwen vision-language model for visual reasoning, documents, and agent tasks"
family = "qwen"
release_date = "2024-01-25"
last_updated = "2025-08-15"
attachment = false
reasoning = false
temperature = true
tool_call = true
knowledge = "2024-04"
open_weights = false
[limit]
context = 131_072
output = 8_192
[modalities]
input = ["text", "image"]
output = ["text"]
models/alibaba/qwen2-5-vl-72b-instruct.toml
+23
View File
@@ -0,0 +1,23 @@
name = "Qwen2.5-VL 72B Instruct"
description = "Qwen vision-language model for visual reasoning, documents, and agent tasks"
family = "qwen"
release_date = "2024-09"
last_updated = "2024-09"
attachment = false
reasoning = false
temperature = true
tool_call = true
knowledge = "2024-04"
open_weights = true
[limit]
context = 131_072
output = 8_192
[modalities]
input = ["text", "image"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/Qwen/Qwen2.5-VL-72B-Instruct"
models/alibaba/qwen3-235b-a22b.toml
+37
View File
@@ -0,0 +1,37 @@
name = "Qwen3 235B-A22B"
description = "Large open Qwen MoE for multilingual reasoning, coding, and tool use"
family = "qwen"
release_date = "2025-04"
last_updated = "2025-04"
attachment = false
reasoning = true
temperature = true
tool_call = true
knowledge = "2025-04"
open_weights = true
[limit]
context = 131_072
output = 16_384
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/Qwen/Qwen3-235B-A22B"
[[benchmarks]]
name = "Aider Polyglot"
score = 59.6
metric = "percent correct"
source = "https://aider.chat/docs/leaderboards/"
date = "2025-05-09"
[[benchmarks]]
name = "SWE-Bench Pro"
score = 21.41
metric = "resolve rate"
dataset = "public"
source = "https://labs.scale.com/leaderboard/swe_bench_pro_public"
models/alibaba/qwen3-32b.toml
+30
View File
@@ -0,0 +1,30 @@
name = "Qwen3 32B"
description = "Dense open Qwen model for self-hosted chat, reasoning, and coding"
family = "qwen"
release_date = "2025-04"
last_updated = "2025-04"
attachment = false
reasoning = true
temperature = true
tool_call = true
knowledge = "2025-04"
open_weights = true
[limit]
context = 131_072
output = 16_384
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/Qwen/Qwen3-32B"
[[benchmarks]]
name = "Aider Polyglot"
score = 40.0
metric = "percent correct"
source = "https://aider.chat/docs/leaderboards/"
date = "2025-05-08"
models/alibaba/qwen3-coder-30b-a3b-instruct.toml
+44
View File
@@ -0,0 +1,44 @@
name = "Qwen3-Coder 30B-A3B Instruct"
description = "Smaller Qwen coder for efficient local agents and repo-level fixes"
family = "qwen"
release_date = "2025-04"
last_updated = "2025-04"
attachment = false
reasoning = false
temperature = true
tool_call = true
knowledge = "2025-04"
open_weights = true
[limit]
context = 262_144
output = 65_536
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/Qwen/Qwen3-Coder-30B-A3B-Instruct"
[[benchmarks]]
name = "Artificial Analysis Coding Index"
score = 19.4
metric = "index"
source = "https://openrouter.ai/qwen/qwen3-coder-30b-a3b-instruct/benchmarks"
date = "2026-06-02"
[[benchmarks]]
name = "SciCode"
score = 27.8
metric = "percent correct"
source = "https://openrouter.ai/qwen/qwen3-coder-30b-a3b-instruct/benchmarks"
date = "2026-06-02"
[[benchmarks]]
name = "Terminal-Bench Hard"
score = 15.2
metric = "success rate"
source = "https://openrouter.ai/qwen/qwen3-coder-30b-a3b-instruct/benchmarks"
date = "2026-06-02"
models/alibaba/qwen3-coder-480b-a35b-instruct.toml
+30
View File
@@ -0,0 +1,30 @@
name = "Qwen3-Coder 480B-A35B Instruct"
description = "Open Qwen coding heavyweight for repository reasoning and agentic engineering"
family = "qwen"
release_date = "2025-04"
last_updated = "2025-04"
attachment = false
reasoning = false
temperature = true
tool_call = true
knowledge = "2025-04"
open_weights = true
[limit]
context = 262_144
output = 65_536
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/Qwen/Qwen3-Coder-480B-A35B-Instruct"
[[benchmarks]]
name = "SWE-Bench Pro"
score = 38.7
metric = "resolve rate"
dataset = "public"
source = "https://labs.scale.com/leaderboard/swe_bench_pro_public"
models/alibaba/qwen3-coder-flash.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Qwen3 Coder Flash"
description = "Qwen coding model for software agents, repository edits, and code reasoning"
family = "qwen"
release_date = "2025-07-28"
last_updated = "2025-07-28"
attachment = false
reasoning = false
temperature = true
tool_call = true
knowledge = "2025-04"
open_weights = false
[limit]
context = 1_000_000
output = 65_536
[modalities]
input = ["text"]
output = ["text"]
models/alibaba/qwen3-coder-plus.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Qwen3 Coder Plus"
description = "Hosted Qwen coder for software agents, repo edits, and long-context code"
family = "qwen"
release_date = "2025-07-23"
last_updated = "2025-07-23"
attachment = false
reasoning = false
temperature = true
tool_call = true
knowledge = "2025-04"
open_weights = false
[limit]
context = 1_048_576
output = 65_536
[modalities]
input = ["text"]
output = ["text"]
models/alibaba/qwen3-max.toml
+40
View File
@@ -0,0 +1,40 @@
name = "Qwen3 Max"
description = "Flagship Qwen3 model for coding agents, complex reasoning, and tool use"
family = "qwen"
release_date = "2025-09-23"
last_updated = "2025-09-23"
attachment = false
reasoning = false
temperature = true
tool_call = true
knowledge = "2025-04"
open_weights = false
[limit]
context = 262_144
output = 65_536
[modalities]
input = ["text"]
output = ["text"]
[[benchmarks]]
name = "Artificial Analysis Coding Index"
score = 26.4
metric = "index"
source = "https://openrouter.ai/qwen/qwen3-max/benchmarks"
date = "2026-05-30"
[[benchmarks]]
name = "SciCode"
score = 38.3
metric = "percent correct"
source = "https://openrouter.ai/qwen/qwen3-max/benchmarks"
date = "2026-05-30"
[[benchmarks]]
name = "Terminal-Bench Hard"
score = 20.5
metric = "success rate"
source = "https://openrouter.ai/qwen/qwen3-max/benchmarks"
date = "2026-05-30"
models/alibaba/qwen3-next-80b-a3b-instruct.toml
+23
View File
@@ -0,0 +1,23 @@
name = "Qwen3-Next 80B-A3B Instruct"
description = "Qwen instruction model for multilingual chat, reasoning, and tool use"
family = "qwen"
release_date = "2025-09"
last_updated = "2025-09"
attachment = false
reasoning = false
temperature = true
tool_call = true
knowledge = "2025-04"
open_weights = true
[limit]
context = 131_072
output = 32_768
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/Qwen/Qwen3-Next-80B-A3B-Instruct"
models/alibaba/qwen3-next-80b-a3b-thinking.toml
+23
View File
@@ -0,0 +1,23 @@
name = "Qwen3-Next 80B-A3B (Thinking)"
description = "Efficient Qwen thinking model for local reasoning, math, and coding agents"
family = "qwen"
release_date = "2025-09"
last_updated = "2025-09"
attachment = false
reasoning = true
temperature = true
tool_call = true
knowledge = "2025-04"
open_weights = true
[limit]
context = 131_072
output = 32_768
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/Qwen/Qwen3-Next-80B-A3B-Thinking"
models/alibaba/qwen3-vl-plus.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Qwen3-VL Plus"
description = "Qwen vision-language model for visual reasoning, documents, and agent tasks"
family = "qwen"
release_date = "2025-09-23"
last_updated = "2025-09-23"
attachment = false
reasoning = true
temperature = true
tool_call = true
knowledge = "2025-04"
open_weights = false
[limit]
context = 262_144
output = 32_768
[modalities]
input = ["text", "image"]
output = ["text"]
models/alibaba/qwen3.5-122b-a10b.toml
+29
View File
@@ -0,0 +1,29 @@
name = "Qwen3.5 122B-A10B"
description = "Qwen vision-language model for visual reasoning, documents, and agent tasks"
family = "qwen"
release_date = "2026-02-23"
last_updated = "2026-02-23"
attachment = true
reasoning = true
temperature = true
tool_call = true
structured_output = true
open_weights = true
[limit]
context = 262_144
output = 65_536
[modalities]
input = ["text", "image", "video", "audio"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/Qwen/Qwen3.5-122B-A10B"
[[benchmarks]]
name = "SWE-Bench Verified"
score = 72
metric = "resolved"
source = "https://huggingface.co/Qwen/Qwen3.5-122B-A10B"
models/alibaba/qwen3.5-27b.toml
+29
View File
@@ -0,0 +1,29 @@
name = "Qwen3.5 27B"
description = "Qwen vision-language model for visual reasoning, documents, and agent tasks"
family = "qwen"
release_date = "2026-02-23"
last_updated = "2026-02-23"
attachment = true
reasoning = true
temperature = true
tool_call = true
structured_output = true
open_weights = true
[limit]
context = 262_144
output = 65_536
[modalities]
input = ["text", "image", "video", "audio"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/Qwen/Qwen3.5-27B"
[[benchmarks]]
name = "SWE-Bench Verified"
score = 72.4
metric = "resolved"
source = "https://huggingface.co/Qwen/Qwen3.5-27B"
models/alibaba/qwen3.5-35b-a3b.toml
+23
View File
@@ -0,0 +1,23 @@
name = "Qwen3.5 35B-A3B"
description = "Qwen vision-language model for visual reasoning, documents, and agent tasks"
family = "qwen"
release_date = "2026-02-23"
last_updated = "2026-02-23"
attachment = true
reasoning = true
temperature = true
tool_call = true
structured_output = true
open_weights = true
[limit]
context = 262_144
output = 65_536
[modalities]
input = ["text", "image", "video", "audio"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/Qwen/Qwen3.5-35B-A3B"
models/alibaba/qwen3.5-397b-a17b.toml
+29
View File
@@ -0,0 +1,29 @@
name = "Qwen3.5 397B-A17B"
description = "Large open Qwen multimodal MoE for visual agents and long technical tasks"
family = "qwen"
release_date = "2026-02-15"
last_updated = "2026-02-15"
attachment = true
reasoning = true
temperature = true
tool_call = true
structured_output = true
open_weights = true
[limit]
context = 262_144
output = 65_536
[modalities]
input = ["text", "image", "video", "audio"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/Qwen/Qwen3.5-397B-A17B"
[[benchmarks]]
name = "SWE-Bench Verified"
score = 76.4
metric = "resolved"
source = "https://huggingface.co/Qwen/Qwen3.5-397B-A17B"
models/alibaba/qwen3.5-9b.toml
+23
View File
@@ -0,0 +1,23 @@
name = "Qwen3.5 9B"
description = "Qwen instruction model for multilingual chat, reasoning, and tool use"
family = "qwen"
release_date = "2026-02-23"
last_updated = "2026-02-23"
attachment = false
reasoning = true
temperature = true
tool_call = true
structured_output = true
open_weights = true
[limit]
context = 262_144
output = 65_536
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/Qwen/Qwen3.5-9B"
models/alibaba/qwen3.5-plus.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Qwen3.5 Plus"
description = "Qwen vision-language model for visual reasoning, documents, and agent tasks"
family = "qwen"
release_date = "2026-02-16"
last_updated = "2026-02-16"
attachment = false
reasoning = true
temperature = true
tool_call = true
knowledge = "2025-04"
open_weights = false
[limit]
context = 1_000_000
output = 65_536
[modalities]
input = ["text", "image", "video"]
output = ["text"]
models/alibaba/qwen3.6-27b.toml
+29
View File
@@ -0,0 +1,29 @@
name = "Qwen3.6 27B"
description = "Qwen vision-language model for visual reasoning, documents, and agent tasks"
family = "qwen"
release_date = "2026-04-22"
last_updated = "2026-04-22"
attachment = true
reasoning = true
temperature = true
tool_call = true
structured_output = true
open_weights = true
[limit]
context = 262_144
output = 65_536
[modalities]
input = ["text", "image", "video", "audio"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/Qwen/Qwen3.6-27B"
[[benchmarks]]
name = "SWE-Bench Verified"
score = 77.2
metric = "resolved"
source = "https://huggingface.co/Qwen/Qwen3.6-27B"
models/alibaba/qwen3.6-35b-a3b.toml
+29
View File
@@ -0,0 +1,29 @@
name = "Qwen3.6 35B-A3B"
description = "Open multimodal Qwen MoE for local agents that need vision, audio, and code"
family = "qwen"
release_date = "2026-04-17"
last_updated = "2026-04-17"
attachment = true
reasoning = true
temperature = true
tool_call = true
structured_output = true
open_weights = true
[limit]
context = 262_144
output = 65_536
[modalities]
input = ["text", "image", "video", "audio"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/Qwen/Qwen3.6-35B-A3B"
[[benchmarks]]
name = "SWE-Bench Verified"
score = 73.4
metric = "resolved"
source = "https://huggingface.co/Qwen/Qwen3.6-35B-A3B"
models/alibaba/qwen3.6-flash.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Qwen3.6 Flash"
description = "Qwen vision-language model for visual reasoning, documents, and agent tasks"
family = "qwen3.6"
release_date = "2026-04-27"
last_updated = "2026-04-27"
attachment = true
reasoning = true
temperature = true
tool_call = true
structured_output = true
open_weights = false
[limit]
context = 1_000_000
output = 65_536
[modalities]
input = ["text", "image", "video"]
output = ["text"]
models/alibaba/qwen3.6-max-preview.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Qwen3.6 Max Preview"
description = "Flagship Qwen model for complex reasoning, coding, and agentic workflows"
family = "qwen"
release_date = "2026-04-20"
last_updated = "2026-04-20"
attachment = false
reasoning = true
temperature = true
tool_call = true
knowledge = "2025-04"
open_weights = false
[limit]
context = 262_144
output = 65_536
[modalities]
input = ["text"]
output = ["text"]
models/alibaba/qwen3.6-plus.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Qwen3.6 Plus"
description = "Earlier Qwen multimodal workhorse for million-token agent and document tasks"
family = "qwen"
release_date = "2026-04-02"
last_updated = "2026-04-02"
attachment = false
reasoning = true
temperature = true
tool_call = true
knowledge = "2025-04"
open_weights = false
[limit]
context = 1_000_000
output = 65_536
[modalities]
input = ["text", "image", "video"]
output = ["text"]
models/alibaba/qwen3.7-max.toml
+82
View File
@@ -0,0 +1,82 @@
name = "Qwen3.7 Max"
description = "Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks"
family = "qwen"
release_date = "2026-05-21"
last_updated = "2026-05-21"
attachment = false
reasoning = true
temperature = true
tool_call = true
open_weights = false
[limit]
context = 1_000_000
output = 65_536
[modalities]
input = ["text"]
output = ["text"]
[[benchmarks]]
name = "SWE-Bench Verified"
score = 80.4
metric = "resolved"
source = "https://qwen.ai/blog?id=qwen3.7"
date = "2026-05-19"
[[benchmarks]]
name = "SWE-Bench Pro"
score = 60.6
metric = "resolve rate"
source = "https://qwen.ai/blog?id=qwen3.7"
date = "2026-05-19"
[[benchmarks]]
name = "SWE-Bench Multilingual"
score = 78.3
metric = "resolve rate"
source = "https://qwen.ai/blog?id=qwen3.7"
date = "2026-05-19"
[[benchmarks]]
name = "Terminal-Bench"
score = 69.7
metric = "success rate"
harness = "Terminus-2"
version = "2.0"
source = "https://qwen.ai/blog?id=qwen3.7"
date = "2026-05-19"
[[benchmarks]]
name = "GPQA Diamond"
score = 92.4
metric = "accuracy"
source = "https://qwen.ai/blog?id=qwen3.7"
date = "2026-05-19"
[[benchmarks]]
name = "Humanity's Last Exam"
score = 41.4
metric = "accuracy"
source = "https://qwen.ai/blog?id=qwen3.7"
date = "2026-05-19"
[[benchmarks]]
name = "SciCode"
score = 53.5
source = "https://qwen.ai/blog?id=qwen3.7"
date = "2026-05-19"
[[benchmarks]]
name = "MCP Atlas"
score = 76.4
metric = "success rate"
source = "https://qwen.ai/blog?id=qwen3.7"
date = "2026-05-19"
[[benchmarks]]
name = "NL2Repo"
score = 47.2
harness = "Claude Code"
source = "https://qwen.ai/blog?id=qwen3.7"
date = "2026-05-19"
models/alibaba/qwen3.7-plus.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Qwen3.7 Plus"
description = "Multimodal Qwen workhorse for long-context agents, visual inputs, and coding"
family = "qwen"
release_date = "2026-06-02"
last_updated = "2026-06-02"
attachment = false
reasoning = true
temperature = true
tool_call = true
knowledge = "2025-04"
open_weights = false
[limit]
context = 1_000_000
output = 64_000
[modalities]
input = ["text", "image"]
output = ["text"]
models/alibaba/qwq-plus.toml
+19
View File
@@ -0,0 +1,19 @@
name = "QwQ Plus"
description = "Qwen reasoning model for deliberate problem solving, math, and coding"
family = "qwen"
release_date = "2025-03-05"
last_updated = "2025-03-05"
attachment = false
reasoning = true
temperature = true
tool_call = true
knowledge = "2024-04"
open_weights = false
[limit]
context = 131_072
output = 8_192
[modalities]
input = ["text"]
output = ["text"]
models/anthropic/claude-3-5-haiku-20241022.toml
+26
View File
@@ -0,0 +1,26 @@
name = "Claude Haiku 3.5"
description = "Fast Claude model for responsive assistance, classification, and lightweight agents"
family = "claude-haiku"
release_date = "2024-10-22"
last_updated = "2024-10-22"
attachment = true
reasoning = false
temperature = true
tool_call = true
knowledge = "2024-07-31"
open_weights = false
[limit]
context = 200_000
output = 8_192
[modalities]
input = ["text", "image", "pdf"]
output = ["text"]
[[benchmarks]]
name = "Aider Polyglot"
score = 28.0
metric = "percent correct"
source = "https://aider.chat/docs/leaderboards/"
date = "2024-12-21"
models/anthropic/claude-3-5-sonnet-20241022.toml
+26
View File
@@ -0,0 +1,26 @@
name = "Claude Sonnet 3.5 v2"
description = "Balanced Claude model for coding, analysis, agent workflows, and cost control"
family = "claude-sonnet"
release_date = "2024-10-22"
last_updated = "2024-10-22"
attachment = true
reasoning = false
temperature = true
tool_call = true
knowledge = "2024-04-30"
open_weights = false
[limit]
context = 200_000
output = 8_192
[modalities]
input = ["text", "image", "pdf"]
output = ["text"]
[[benchmarks]]
name = "Aider Polyglot"
score = 51.6
metric = "percent correct"
source = "https://aider.chat/docs/leaderboards/"
date = "2025-01-17"
models/anthropic/claude-3-7-sonnet-20250219.toml
+26
View File
@@ -0,0 +1,26 @@
name = "Claude Sonnet 3.7"
description = "Balanced Claude model for coding, analysis, agent workflows, and cost control"
family = "claude-sonnet"
release_date = "2025-02-19"
last_updated = "2025-02-19"
attachment = true
reasoning = true
temperature = true
tool_call = true
knowledge = "2024-10-31"
open_weights = false
[limit]
context = 200_000
output = 64_000
[modalities]
input = ["text", "image", "pdf"]
output = ["text"]
[[benchmarks]]
name = "Aider Polyglot"
score = 64.9
metric = "percent correct"
source = "https://aider.chat/docs/leaderboards/"
date = "2025-02-24"
models/anthropic/claude-3-haiku-20240307.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Claude Haiku 3"
description = "Legacy model retained for compatibility with older integrations"
family = "claude-haiku"
release_date = "2024-03-13"
last_updated = "2024-03-13"
attachment = true
reasoning = false
temperature = true
tool_call = true
knowledge = "2023-08-31"
open_weights = false
[limit]
context = 200_000
output = 4_096
[modalities]
input = ["text", "image", "pdf"]
output = ["text"]
models/anthropic/claude-fable-5.toml
+86
View File
@@ -0,0 +1,86 @@
name = "Claude Fable 5"
description = "Claude model for creative writing, analysis, and controlled agent workflows"
family = "claude-fable"
release_date = "2026-06-09"
last_updated = "2026-06-09"
attachment = true
reasoning = true
temperature = false
tool_call = true
open_weights = false
knowledge = "2026-01-31"
[limit]
context = 1_000_000
output = 128_000
[modalities]
input = ["text", "image", "pdf"]
output = ["text"]
[[benchmarks]]
name = "SWE-Bench Pro"
score = 80.3
metric = "resolve rate"
source = "https://www.anthropic.com/news/claude-fable-5-mythos-5"
date = "2026-06-09"
[[benchmarks]]
name = "SWE-Bench Verified"
score = 95
metric = "resolved"
source = "https://benchlm.ai/benchmarks/sweVerified"
[[benchmarks]]
name = "Terminal-Bench"
score = 88.0
metric = "success rate"
version = "2.1"
source = "https://www.anthropic.com/news/claude-fable-5-mythos-5"
date = "2026-06-09"
[[benchmarks]]
name = "Humanity's Last Exam"
score = 59
metric = "accuracy"
variant = "no tools"
source = "https://www.anthropic.com/news/claude-fable-5-mythos-5"
date = "2026-06-09"
[[benchmarks]]
name = "Humanity's Last Exam"
score = 64.5
metric = "accuracy"
variant = "with tools"
source = "https://www.anthropic.com/news/claude-fable-5-mythos-5"
date = "2026-06-09"
[[benchmarks]]
name = "OSWorld-Verified"
score = 85
metric = "success rate"
source = "https://www.anthropic.com/news/claude-fable-5-mythos-5"
date = "2026-06-09"
[[benchmarks]]
name = "FrontierCode"
score = 29.3
metric = "pass rate"
variant = "high effort"
dataset = "Diamond"
source = "https://www.anthropic.com/news/claude-fable-5-mythos-5"
date = "2026-06-09"
[[benchmarks]]
name = "GDPval-AA"
score = 1932
metric = "Elo"
source = "https://www.anthropic.com/news/claude-fable-5-mythos-5"
date = "2026-06-09"
[[benchmarks]]
name = "AutomationBench"
score = 17.4
metric = "success rate"
source = "https://www.anthropic.com/news/claude-fable-5-mythos-5"
date = "2026-06-09"
models/anthropic/claude-haiku-4-5-20251001.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Claude Haiku 4.5"
description = "Fast Claude model for responsive assistance, classification, and lightweight agents"
family = "claude-haiku"
release_date = "2025-10-15"
last_updated = "2025-10-15"
attachment = true
reasoning = true
temperature = true
tool_call = true
knowledge = "2025-02-28"
open_weights = false
[limit]
context = 200_000
output = 64_000
[modalities]
input = ["text", "image", "pdf"]
output = ["text"]
models/anthropic/claude-haiku-4-5.toml
+26
View File
@@ -0,0 +1,26 @@
name = "Claude Haiku 4.5 (latest)"
description = "Fast Claude lane for lightweight agents, office tasks, and responsive chat"
family = "claude-haiku"
release_date = "2025-10-15"
last_updated = "2025-10-15"
attachment = true
reasoning = true
temperature = true
tool_call = true
knowledge = "2025-02-28"
open_weights = false
[limit]
context = 200_000
output = 64_000
[modalities]
input = ["text", "image", "pdf"]
output = ["text"]
[[benchmarks]]
name = "SWE-Bench Pro"
score = 39.45
metric = "resolve rate"
dataset = "public"
source = "https://labs.scale.com/leaderboard/swe_bench_pro_public"
models/anthropic/claude-opus-4-0.toml
+26
View File
@@ -0,0 +1,26 @@
name = "Claude Opus 4 (latest)"
description = "Flagship Claude model for deep reasoning, coding, and long-horizon agents"
family = "claude-opus"
release_date = "2025-05-22"
last_updated = "2025-05-22"
attachment = true
reasoning = true
temperature = true
tool_call = true
knowledge = "2025-03-31"
open_weights = false
[limit]
context = 200_000
output = 32_000
[modalities]
input = ["text", "image", "pdf"]
output = ["text"]
[[benchmarks]]
name = "Aider Polyglot"
score = 72.0
metric = "percent correct"
source = "https://aider.chat/docs/leaderboards/"
date = "2025-05-25"
models/anthropic/claude-opus-4-1-20250805.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Claude Opus 4.1"
description = "Flagship Claude model for deep reasoning, coding, and long-horizon agents"
family = "claude-opus"
release_date = "2025-08-05"
last_updated = "2025-08-05"
attachment = true
reasoning = true
temperature = true
tool_call = true
knowledge = "2025-03-31"
open_weights = false
[limit]
context = 200_000
output = 32_000
[modalities]
input = ["text", "image", "pdf"]
output = ["text"]
models/anthropic/claude-opus-4-1.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Claude Opus 4.1 (latest)"
description = "Flagship Claude model for deep reasoning, coding, and long-horizon agents"
family = "claude-opus"
release_date = "2025-08-05"
last_updated = "2025-08-05"
attachment = true
reasoning = true
temperature = true
tool_call = true
knowledge = "2025-03-31"
open_weights = false
[limit]
context = 200_000
output = 32_000
[modalities]
input = ["text", "image", "pdf"]
output = ["text"]
models/anthropic/claude-opus-4-20250514.toml
+26
View File
@@ -0,0 +1,26 @@
name = "Claude Opus 4"
description = "Flagship Claude model for deep reasoning, coding, and long-horizon agents"
family = "claude-opus"
release_date = "2025-05-22"
last_updated = "2025-05-22"
attachment = true
reasoning = true
temperature = true
tool_call = true
knowledge = "2025-03-31"
open_weights = false
[limit]
context = 200_000
output = 32_000
[modalities]
input = ["text", "image", "pdf"]
output = ["text"]
[[benchmarks]]
name = "Aider Polyglot"
score = 72.0
metric = "percent correct"
source = "https://aider.chat/docs/leaderboards/"
date = "2025-05-25"
models/anthropic/claude-opus-4-5-20251101.toml
+26
View File
@@ -0,0 +1,26 @@
name = "Claude Opus 4.5"
description = "Flagship Claude model for deep reasoning, coding, and long-horizon agents"
family = "claude-opus"
release_date = "2025-11-01"
last_updated = "2025-11-01"
attachment = true
reasoning = true
temperature = true
tool_call = true
knowledge = "2025-05"
open_weights = false
[limit]
context = 200_000
output = 64_000
[modalities]
input = ["text", "image", "pdf"]
output = ["text"]
[[benchmarks]]
name = "SWE-Bench Pro"
score = 45.89
metric = "resolve rate"
dataset = "public"
source = "https://labs.scale.com/leaderboard/swe_bench_pro_public"
models/anthropic/claude-opus-4-5.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Claude Opus 4.5 (latest)"
description = "Flagship Claude model for deep reasoning, coding, and long-horizon agents"
family = "claude-opus"
release_date = "2025-11-24"
last_updated = "2025-11-24"
attachment = true
reasoning = true
temperature = true
tool_call = true
knowledge = "2025-05"
open_weights = false
[limit]
context = 200_000
output = 64_000
[modalities]
input = ["text", "image", "pdf"]
output = ["text"]
models/anthropic/claude-opus-4-6.toml
+95
View File
@@ -0,0 +1,95 @@
name = "Claude Opus 4.6"
description = "High-end Claude for difficult coding, planning, and slower expert reasoning"
family = "claude-opus"
release_date = "2026-02-05"
last_updated = "2026-03-13"
attachment = true
reasoning = true
temperature = true
tool_call = true
knowledge = "2025-05-31"
open_weights = false
[limit]
context = 1_000_000
output = 128_000
[modalities]
input = ["text", "image", "pdf"]
output = ["text"]
[[benchmarks]]
name = "SWE-Bench Pro"
score = 51.9
metric = "resolve rate"
dataset = "public"
source = "https://labs.scale.com/leaderboard/swe_bench_pro_public"
[[benchmarks]]
name = "SWE-Atlas Codebase QnA"
score = 33.3
metric = "score"
harness = "Claude Code"
source = "https://labs.scale.com/leaderboard/sweatlas-qna"
[[benchmarks]]
name = "SWE-Atlas Codebase QnA"
score = 30
metric = "score"
harness = "Mini-SWE-Agent"
source = "https://labs.scale.com/leaderboard/sweatlas-qna"
[[benchmarks]]
name = "SWE-Atlas Refactoring"
score = 35.58
metric = "score"
harness = "Claude Code"
source = "https://labs.scale.com/leaderboard/sweatlas-refactoring"
[[benchmarks]]
name = "SWE-Atlas Test Writing"
score = 36.67
metric = "score"
harness = "Claude Code"
source = "https://labs.scale.com/leaderboard/sweatlas-tw"
[[benchmarks]]
name = "SWE-Atlas Test Writing"
score = 36.08
metric = "score"
harness = "Mini-SWE-Agent"
source = "https://labs.scale.com/leaderboard/sweatlas-tw"
[[benchmarks]]
name = "Artificial Analysis Coding Agent Index"
score = 51.3
metric = "average pass@1"
harness = "Claude Code"
variant = "medium"
source = "https://artificialanalysis.ai/agents/coding-agents"
[[benchmarks]]
name = "SWE-Atlas Codebase QnA"
score = 71.9
metric = "pass@1"
harness = "Claude Code"
variant = "medium"
source = "https://artificialanalysis.ai/agents/coding-agents"
[[benchmarks]]
name = "SWE-Bench Pro"
score = 11.8
metric = "pass@1"
harness = "Claude Code"
variant = "medium"
dataset = "hard-aa"
source = "https://artificialanalysis.ai/agents/coding-agents"
[[benchmarks]]
name = "Terminal-Bench"
score = 70.2
metric = "pass@1"
harness = "Claude Code"
variant = "medium"
version = "2.1"
source = "https://artificialanalysis.ai/agents/coding-agents"
models/anthropic/claude-opus-4-7.toml
+174
View File
@@ -0,0 +1,174 @@
name = "Claude Opus 4.7"
description = "Stronger Opus tier for advanced software work and high-stakes reasoning"
family = "claude-opus"
release_date = "2026-04-16"
last_updated = "2026-04-16"
attachment = true
reasoning = true
temperature = false
tool_call = true
knowledge = "2026-01-31"
open_weights = false
[limit]
context = 1_000_000
output = 128_000
[modalities]
input = ["text", "image", "pdf"]
output = ["text"]
[[benchmarks]]
name = "SWE-Bench Pro"
score = 64.3
metric = "resolve rate"
source = "https://www.anthropic.com/news/claude-opus-4-8"
date = "2026-05-28"
[[benchmarks]]
name = "Terminal-Bench"
score = 66.1
metric = "success rate"
harness = "Terminus-2"
version = "2.1"
source = "https://www.anthropic.com/news/claude-opus-4-8"
date = "2026-05-28"
[[benchmarks]]
name = "SWE-Atlas Refactoring"
score = 48.57
metric = "score"
harness = "Claude Code"
source = "https://labs.scale.com/leaderboard/sweatlas-refactoring"
[[benchmarks]]
name = "Artificial Analysis Coding Agent Index"
score = 66.6
metric = "average pass@1"
harness = "Claude Code"
variant = "max"
source = "https://artificialanalysis.ai/agents/coding-agents"
[[benchmarks]]
name = "SWE-Atlas Codebase QnA"
score = 81
metric = "pass@1"
harness = "Claude Code"
variant = "max"
source = "https://artificialanalysis.ai/agents/coding-agents"
[[benchmarks]]
name = "SWE-Bench Pro"
score = 44.9
metric = "pass@1"
harness = "Claude Code"
variant = "max"
dataset = "hard-aa"
source = "https://artificialanalysis.ai/agents/coding-agents"
[[benchmarks]]
name = "Terminal-Bench"
score = 73.8
metric = "pass@1"
harness = "Claude Code"
variant = "max"
version = "2.1"
source = "https://artificialanalysis.ai/agents/coding-agents"
[[benchmarks]]
name = "Artificial Analysis Coding Agent Index"
score = 61.2
metric = "average pass@1"
harness = "Cursor CLI"
variant = "medium"
source = "https://artificialanalysis.ai/agents/coding-agents"
[[benchmarks]]
name = "SWE-Atlas Codebase QnA"
score = 78.4
metric = "pass@1"
harness = "Cursor CLI"
variant = "medium"
source = "https://artificialanalysis.ai/agents/coding-agents"
[[benchmarks]]
name = "SWE-Bench Pro"
score = 34.4
metric = "pass@1"
harness = "Cursor CLI"
variant = "medium"
dataset = "hard-aa"
source = "https://artificialanalysis.ai/agents/coding-agents"
[[benchmarks]]
name = "Terminal-Bench"
score = 70.6
metric = "pass@1"
harness = "Cursor CLI"
variant = "medium"
version = "2.1"
source = "https://artificialanalysis.ai/agents/coding-agents"
[[benchmarks]]
name = "Artificial Analysis Coding Agent Index"
score = 59.9
metric = "average pass@1"
harness = "Claude Code"
variant = "medium"
source = "https://artificialanalysis.ai/agents/coding-agents"
[[benchmarks]]
name = "SWE-Atlas Codebase QnA"
score = 71.7
metric = "pass@1"
harness = "Claude Code"
variant = "medium"
source = "https://artificialanalysis.ai/agents/coding-agents"
[[benchmarks]]
name = "SWE-Bench Pro"
score = 36.4
metric = "pass@1"
harness = "Claude Code"
variant = "medium"
dataset = "hard-aa"
source = "https://artificialanalysis.ai/agents/coding-agents"
[[benchmarks]]
name = "Terminal-Bench"
score = 71.4
metric = "pass@1"
harness = "Claude Code"
variant = "medium"
version = "2.1"
source = "https://artificialanalysis.ai/agents/coding-agents"
[[benchmarks]]
name = "GPQA Diamond"
score = 94.2
metric = "accuracy"
source = "https://openai.com/index/introducing-gpt-5-5/"
date = "2026-04-23"
[[benchmarks]]
name = "Humanity's Last Exam"
score = 46.9
metric = "accuracy"
variant = "no tools"
source = "https://openai.com/index/introducing-gpt-5-5/"
date = "2026-04-23"
[[benchmarks]]
name = "Humanity's Last Exam"
score = 54.7
metric = "accuracy"
variant = "with tools"
source = "https://openai.com/index/introducing-gpt-5-5/"
date = "2026-04-23"
[[benchmarks]]
name = "OSWorld-Verified"
score = 78.0
metric = "success rate"
source = "https://openai.com/index/introducing-gpt-5-5/"
date = "2026-04-23"
models/anthropic/claude-opus-4-8.toml
+73
View File
@@ -0,0 +1,73 @@
name = "Claude Opus 4.8"
description = "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents"
family = "claude-opus"
release_date = "2026-05-28"
last_updated = "2026-05-28"
attachment = true
reasoning = true
temperature = false
tool_call = true
open_weights = false
knowledge = "2026-01"
[limit]
context = 1_000_000
output = 128_000
[modalities]
input = ["text", "image", "pdf"]
output = ["text"]
[[benchmarks]]
name = "SWE-Bench Pro"
score = 69.2
metric = "resolve rate"
source = "https://www.anthropic.com/news/claude-opus-4-8"
date = "2026-05-28"
[[benchmarks]]
name = "Terminal-Bench"
score = 74.6
metric = "success rate"
harness = "Terminus-2"
version = "2.1"
source = "https://www.anthropic.com/news/claude-opus-4-8"
date = "2026-05-28"
[[benchmarks]]
name = "SWE-Bench Verified"
score = 88.6
metric = "resolved"
source = "https://benchlm.ai/benchmarks/sweVerified"
[[benchmarks]]
name = "Humanity's Last Exam"
score = 49.8
metric = "accuracy"
variant = "no tools"
source = "https://www.anthropic.com/news/claude-fable-5-mythos-5"
date = "2026-06-09"
[[benchmarks]]
name = "Humanity's Last Exam"
score = 57.9
metric = "accuracy"
variant = "with tools"
source = "https://www.anthropic.com/news/claude-fable-5-mythos-5"
date = "2026-06-09"
[[benchmarks]]
name = "OSWorld-Verified"
score = 83.4
metric = "success rate"
source = "https://www.anthropic.com/news/claude-fable-5-mythos-5"
date = "2026-06-09"
[[benchmarks]]
name = "FrontierCode"
score = 13.4
metric = "pass rate"
variant = "high effort"
dataset = "Diamond"
source = "https://www.anthropic.com/news/claude-fable-5-mythos-5"
date = "2026-06-09"
models/anthropic/claude-sonnet-4-0.toml
+33
View File
@@ -0,0 +1,33 @@
name = "Claude Sonnet 4 (latest)"
description = "Balanced Claude model for coding, analysis, agent workflows, and cost control"
family = "claude-sonnet"
release_date = "2025-05-22"
last_updated = "2025-05-22"
attachment = true
reasoning = true
temperature = true
tool_call = true
knowledge = "2025-03-31"
open_weights = false
[limit]
context = 200_000
output = 64_000
[modalities]
input = ["text", "image", "pdf"]
output = ["text"]
[[benchmarks]]
name = "Aider Polyglot"
score = 61.3
metric = "percent correct"
source = "https://aider.chat/docs/leaderboards/"
date = "2025-05-24"
[[benchmarks]]
name = "SWE-Bench Pro"
score = 42.7
metric = "resolve rate"
dataset = "public"
source = "https://labs.scale.com/leaderboard/swe_bench_pro_public"
models/anthropic/claude-sonnet-4-20250514.toml
+26
View File
@@ -0,0 +1,26 @@
name = "Claude Sonnet 4"
description = "Balanced Claude model for coding, analysis, agent workflows, and cost control"
family = "claude-sonnet"
release_date = "2025-05-22"
last_updated = "2025-05-22"
attachment = true
reasoning = true
temperature = true
tool_call = true
knowledge = "2025-03-31"
open_weights = false
[limit]
context = 200_000
output = 64_000
[modalities]
input = ["text", "image", "pdf"]
output = ["text"]
[[benchmarks]]
name = "Aider Polyglot"
score = 61.3
metric = "percent correct"
source = "https://aider.chat/docs/leaderboards/"
date = "2025-05-24"
models/anthropic/claude-sonnet-4-5-20250929.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Claude Sonnet 4.5"
description = "Balanced Claude model for coding, analysis, agent workflows, and cost control"
family = "claude-sonnet"
release_date = "2025-09-29"
last_updated = "2025-09-29"
attachment = true
reasoning = true
temperature = true
tool_call = true
knowledge = "2025-07-31"
open_weights = false
[limit]
context = 200_000
output = 64_000
[modalities]
input = ["text", "image", "pdf"]
output = ["text"]
models/anthropic/claude-sonnet-4-5.toml
+26
View File
@@ -0,0 +1,26 @@
name = "Claude Sonnet 4.5 (latest)"
description = "Balanced Claude model for coding, analysis, agent workflows, and cost control"
family = "claude-sonnet"
release_date = "2025-09-29"
last_updated = "2025-09-29"
attachment = true
reasoning = true
temperature = true
tool_call = true
knowledge = "2025-07-31"
open_weights = false
[limit]
context = 200_000
output = 64_000
[modalities]
input = ["text", "image", "pdf"]
output = ["text"]
[[benchmarks]]
name = "SWE-Bench Pro"
score = 43.6
metric = "resolve rate"
dataset = "public"
source = "https://labs.scale.com/leaderboard/swe_bench_pro_public"
models/anthropic/claude-sonnet-4-6.toml
+106
View File
@@ -0,0 +1,106 @@
name = "Claude Sonnet 4.6"
description = "Claude workhorse for coding agents, careful analysis, and production cost control"
family = "claude-sonnet"
release_date = "2026-02-17"
last_updated = "2026-03-13"
attachment = true
reasoning = true
temperature = true
tool_call = true
knowledge = "2025-08-31"
open_weights = false
[limit]
context = 1_000_000
output = 64_000
[modalities]
input = ["text", "image", "pdf"]
output = ["text"]
[[benchmarks]]
name = "SWE-Atlas Codebase QnA"
score = 31.2
metric = "score"
harness = "Claude Code"
source = "https://labs.scale.com/leaderboard/sweatlas-qna"
[[benchmarks]]
name = "SWE-Atlas Refactoring"
score = 32.21
metric = "score"
harness = "Claude Code"
source = "https://labs.scale.com/leaderboard/sweatlas-refactoring"
[[benchmarks]]
name = "SWE-Atlas Test Writing"
score = 31.76
metric = "score"
harness = "Claude Code"
source = "https://labs.scale.com/leaderboard/sweatlas-tw"
[[benchmarks]]
name = "Artificial Analysis Coding Agent Index"
score = 49.4
metric = "average pass@1"
harness = "Claude Code"
variant = "medium"
source = "https://artificialanalysis.ai/agents/coding-agents"
[[benchmarks]]
name = "SWE-Atlas Codebase QnA"
score = 70.3
metric = "pass@1"
harness = "Claude Code"
variant = "medium"
source = "https://artificialanalysis.ai/agents/coding-agents"
[[benchmarks]]
name = "SWE-Bench Pro"
score = 14.9
metric = "pass@1"
harness = "Claude Code"
variant = "medium"
dataset = "hard-aa"
source = "https://artificialanalysis.ai/agents/coding-agents"
[[benchmarks]]
name = "Terminal-Bench"
score = 63.1
metric = "pass@1"
harness = "Claude Code"
variant = "medium"
version = "2.1"
source = "https://artificialanalysis.ai/agents/coding-agents"
[[benchmarks]]
name = "Terminal-Bench"
score = 67.0
metric = "success rate"
harness = "Terminus-2"
version = "2.1"
source = "https://www.anthropic.com/news/claude-sonnet-5"
date = "2026-06-30"
[[benchmarks]]
name = "Humanity's Last Exam"
score = 34.6
metric = "accuracy"
variant = "no tools"
source = "https://www.anthropic.com/news/claude-sonnet-5"
date = "2026-06-30"
[[benchmarks]]
name = "Humanity's Last Exam"
score = 46.8
metric = "accuracy"
variant = "with tools"
source = "https://www.anthropic.com/news/claude-sonnet-5"
date = "2026-06-30"
[[benchmarks]]
name = "OSWorld-Verified"
score = 78.5
metric = "success rate"
source = "https://www.anthropic.com/news/claude-sonnet-5"
date = "2026-06-30"
models/anthropic/claude-sonnet-5.toml
+72
View File
@@ -0,0 +1,72 @@
name = "Claude Sonnet 5"
description = "Everyday Claude agent model for coding, planning, browsing, and general work"
family = "claude-sonnet"
release_date = "2026-06-30"
last_updated = "2026-06-30"
attachment = true
reasoning = true
temperature = false
tool_call = true
knowledge = "2026-01-31"
open_weights = false
[limit]
context = 1_000_000
output = 128_000
[modalities]
input = ["text", "image", "pdf"]
output = ["text"]
[[benchmarks]]
name = "SWE-Bench Verified"
score = 85.2
metric = "resolved"
source = "https://www.anthropic.com/news/claude-sonnet-5"
date = "2026-06-30"
[[benchmarks]]
name = "SWE-Bench Pro"
score = 63.2
metric = "resolve rate"
source = "https://www.anthropic.com/news/claude-sonnet-5"
date = "2026-06-30"
[[benchmarks]]
name = "SWE-Bench Multilingual"
score = 78.3
metric = "resolve rate"
source = "https://www.anthropic.com/news/claude-sonnet-5"
date = "2026-06-30"
[[benchmarks]]
name = "Terminal-Bench"
score = 80.4
metric = "success rate"
harness = "Terminus-2"
version = "2.1"
source = "https://www.anthropic.com/news/claude-sonnet-5"
date = "2026-06-30"
[[benchmarks]]
name = "OSWorld-Verified"
score = 81.2
metric = "success rate"
source = "https://www.anthropic.com/news/claude-sonnet-5"
date = "2026-06-30"
[[benchmarks]]
name = "BrowseComp"
score = 84.7
metric = "accuracy"
variant = "single agent"
source = "https://www.anthropic.com/news/claude-sonnet-5"
date = "2026-06-30"
[[benchmarks]]
name = "FrontierCode"
score = 38.8
metric = "pass rate"
version = "v1"
source = "https://www.anthropic.com/news/claude-sonnet-5"
date = "2026-06-30"
models/cohere/command-a-03-2025.toml
+30
View File
@@ -0,0 +1,30 @@
name = "Command A"
description = "Cohere command model for multilingual enterprise agents, tools, and chat"
family = "command-a"
release_date = "2025-03-13"
last_updated = "2025-03-13"
attachment = false
reasoning = false
temperature = true
tool_call = true
knowledge = "2024-06-01"
open_weights = true
[limit]
context = 256_000
output = 8_000
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/CohereLabs/c4ai-command-a-03-2025"
[[benchmarks]]
name = "Aider Polyglot"
score = 12.0
metric = "percent correct"
source = "https://aider.chat/docs/leaderboards/"
date = "2025-03-14"
models/cohere/command-a-plus-05-2026.toml
+20
View File
@@ -0,0 +1,20 @@
name = "Command A Plus"
description = "Cohere's stronger command model for multilingual agents and enterprise workflows"
family = "command-a"
release_date = "2026-05-20"
last_updated = "2026-06-09"
attachment = true
reasoning = true
temperature = true
knowledge = "2025-04-01"
tool_call = true
open_weights = true
structured_output = true
[limit]
context = 128_000
output = 64_000
[modalities]
input = ["text", "image"]
output = ["text"]
models/cohere/command-r-08-2024.toml
+23
View File
@@ -0,0 +1,23 @@
name = "Command R"
description = "Cohere retrieval model for long-context chat and enterprise RAG workflows"
family = "command-r"
release_date = "2024-08-30"
last_updated = "2024-08-30"
attachment = false
reasoning = false
temperature = true
tool_call = true
knowledge = "2024-06-01"
open_weights = true
[limit]
context = 128_000
output = 4_000
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/CohereLabs/c4ai-command-r-08-2024"
models/cohere/command-r-plus-08-2024.toml
+23
View File
@@ -0,0 +1,23 @@
name = "Command R+"
description = "Cohere's RAG workhorse for long-context enterprise search and tool use"
family = "command-r"
release_date = "2024-08-30"
last_updated = "2024-08-30"
attachment = false
reasoning = false
temperature = true
tool_call = true
knowledge = "2024-06-01"
open_weights = true
[limit]
context = 128_000
output = 4_000
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/CohereLabs/c4ai-command-r-plus-08-2024"
models/cohere/command-r7b-12-2024.toml
+23
View File
@@ -0,0 +1,23 @@
name = "Command R7B"
description = "Cohere retrieval model for long-context chat and enterprise RAG workflows"
family = "command-r"
release_date = "2024-12-02"
last_updated = "2024-12-02"
attachment = false
reasoning = false
temperature = true
tool_call = true
knowledge = "2024-06-01"
open_weights = true
[limit]
context = 128_000
output = 4_000
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/CohereLabs/c4ai-command-r7b-12-2024"
models/cohere/north-mini-code-1-0.toml
+64
View File
@@ -0,0 +1,64 @@
name = "North Mini Code"
description = "Cohere coding model for practical software engineering and agentic edits"
family = "north"
release_date = "2026-06-09"
last_updated = "2026-06-09"
attachment = false
reasoning = true
temperature = true
structured_output = true
knowledge = "2025-09-23"
tool_call = true
open_weights = true
[limit]
context = 256_000
output = 64_000
[modalities]
input = ["text"]
output = ["text"]
[[benchmarks]]
name = "SWE-Bench Verified"
score = 67.6
metric = "resolved"
harness = "SWE-agent"
source = "https://huggingface.co/CohereLabs/North-Mini-Code-1.0"
date = "2026-06-09"
[[benchmarks]]
name = "SWE-Bench Pro"
score = 40.2
metric = "resolve rate"
harness = "SWE-agent"
source = "https://huggingface.co/CohereLabs/North-Mini-Code-1.0"
date = "2026-06-09"
[[benchmarks]]
name = "Artificial Analysis Intelligence Index"
score = 27.6
metric = "index score"
source = "https://artificialanalysis.ai/articles/north-mini-code-cohere-s-small-coding-focused-moe-model"
date = "2026-06-09"
[[benchmarks]]
name = "Artificial Analysis Coding Index"
score = 33.4
metric = "index score"
source = "https://artificialanalysis.ai/articles/north-mini-code-cohere-s-small-coding-focused-moe-model"
date = "2026-06-09"
[[benchmarks]]
name = "GDPval-AA"
score = 14
metric = "win rate"
source = "https://artificialanalysis.ai/articles/north-mini-code-cohere-s-small-coding-focused-moe-model"
date = "2026-06-09"
[[benchmarks]]
name = "τ²-Bench Telecom"
score = 37
metric = "success rate"
source = "https://artificialanalysis.ai/articles/north-mini-code-cohere-s-small-coding-focused-moe-model"
date = "2026-06-09"
models/deepreinforce/ornith-1.0-31b.toml
+28
View File
@@ -0,0 +1,28 @@
# Announced in the Ornith 1.0 family but not yet published on Hugging Face as
# of 2026-06-28 — no weights URL or benchmark scores available yet. Modalities
# and context window are provisional, assumed consistent with the rest of the
# family pending the public release.
# https://deep-reinforce.com/ornith_1_0.html
name = "Ornith 1.0 31B"
description = "Open coding-reasoning model for repository tasks and self-improving agents"
family = "ornith"
release_date = "2026-06-25"
last_updated = "2026-06-25"
attachment = true
reasoning = true
temperature = true
tool_call = true
open_weights = true
license = "MIT"
[limit]
context = 262_144
[modalities]
input = ["text", "image"]
output = ["text"]
[[links]]
label = "Announcement"
url = "https://deep-reinforce.com/ornith_1_0.html"
type = "announcement"
models/deepreinforce/ornith-1.0-35b.toml
+76
View File
@@ -0,0 +1,76 @@
name = "Ornith 1.0 35B"
description = "Large coding-reasoning model for agentic software tasks and RL search"
family = "ornith"
release_date = "2026-06-25"
last_updated = "2026-06-25"
attachment = true
reasoning = true
temperature = true
tool_call = true
open_weights = true
license = "MIT"
[limit]
context = 262_144
[modalities]
input = ["text", "image"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-35B"
[[links]]
label = "Model card"
url = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-35B"
type = "model_card"
[[links]]
label = "Announcement"
url = "https://deep-reinforce.com/ornith_1_0.html"
type = "announcement"
[[benchmarks]]
name = "SWE-Bench Verified"
score = 75.6
metric = "percent resolved"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-35B"
[[benchmarks]]
name = "SWE-Bench Pro"
score = 50.4
metric = "percent resolved"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-35B"
[[benchmarks]]
name = "SWE-Bench Multilingual"
score = 69.3
metric = "percent resolved"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-35B"
[[benchmarks]]
name = "Terminal-Bench 2.1"
score = 64.2
metric = "percent"
variant = "Terminus-2"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-35B"
[[benchmarks]]
name = "Terminal-Bench 2.1"
score = 62.8
metric = "percent"
variant = "Claude Code"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-35B"
[[benchmarks]]
name = "NL2Repo"
score = 34.6
metric = "percent"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-35B"
[[benchmarks]]
name = "Claw-eval"
score = 69.8
metric = "percent"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-35B"
models/deepreinforce/ornith-1.0-397b.toml
+81
View File
@@ -0,0 +1,81 @@
name = "Ornith 1.0 397B"
description = "Large coding-reasoning model for agentic software tasks and RL search"
family = "ornith"
release_date = "2026-06-25"
last_updated = "2026-06-25"
attachment = true
reasoning = true
temperature = true
tool_call = true
open_weights = true
license = "MIT"
[limit]
context = 262_144
[modalities]
input = ["text", "image"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-397B"
[[weights]]
label = "Hugging Face (FP8)"
url = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-397B-FP8"
quantization = "fp8"
[[links]]
label = "Model card"
url = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-397B"
type = "model_card"
[[links]]
label = "Announcement"
url = "https://deep-reinforce.com/ornith_1_0.html"
type = "announcement"
[[benchmarks]]
name = "SWE-Bench Verified"
score = 82.4
metric = "percent resolved"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-397B"
[[benchmarks]]
name = "SWE-Bench Pro"
score = 62.2
metric = "percent resolved"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-397B"
[[benchmarks]]
name = "SWE-Bench Multilingual"
score = 78.9
metric = "percent resolved"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-397B"
[[benchmarks]]
name = "Terminal-Bench 2.1"
score = 77.5
metric = "percent"
variant = "Terminus-2"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-397B"
[[benchmarks]]
name = "Terminal-Bench 2.1"
score = 78.2
metric = "percent"
variant = "Claude Code"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-397B"
[[benchmarks]]
name = "NL2Repo"
score = 48.2
metric = "percent"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-397B"
[[benchmarks]]
name = "Claw-eval"
score = 77.1
metric = "percent"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-397B"
models/deepreinforce/ornith-1.0-9b.toml
+76
View File
@@ -0,0 +1,76 @@
name = "Ornith 1.0 9B"
description = "Open coding-reasoning model for repository tasks and self-improving agents"
family = "ornith"
release_date = "2026-06-25"
last_updated = "2026-06-25"
attachment = true
reasoning = true
temperature = true
tool_call = true
open_weights = true
license = "MIT"
[limit]
context = 262_144
[modalities]
input = ["text", "image"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-9B"
[[links]]
label = "Model card"
url = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-9B"
type = "model_card"
[[links]]
label = "Announcement"
url = "https://deep-reinforce.com/ornith_1_0.html"
type = "announcement"
[[benchmarks]]
name = "SWE-Bench Verified"
score = 69.4
metric = "percent resolved"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-9B"
[[benchmarks]]
name = "SWE-Bench Pro"
score = 42.9
metric = "percent resolved"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-9B"
[[benchmarks]]
name = "SWE-Bench Multilingual"
score = 52
metric = "percent resolved"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-9B"
[[benchmarks]]
name = "Terminal-Bench 2.1"
score = 43.1
metric = "percent"
variant = "Terminus-2"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-9B"
[[benchmarks]]
name = "Terminal-Bench 2.1"
score = 40.6
metric = "percent"
variant = "Claude Code"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-9B"
[[benchmarks]]
name = "NL2Repo"
score = 27.2
metric = "percent"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-9B"
[[benchmarks]]
name = "Claw-eval"
score = 63.1
metric = "percent"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-9B"
models/deepseek/deepseek-chat.toml
+30
View File
@@ -0,0 +1,30 @@
name = "DeepSeek Chat"
description = "DeepSeek chat model for instruction following, coding, and analysis"
family = "deepseek"
release_date = "2025-12-01"
last_updated = "2026-02-28"
attachment = true
reasoning = false
temperature = true
tool_call = true
knowledge = "2025-09"
open_weights = true
[limit]
context = 1_000_000
output = 384_000
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/deepseek-ai/DeepSeek-V3.2"
[[benchmarks]]
name = "Aider Polyglot"
score = 70.2
metric = "percent correct"
source = "https://aider.chat/docs/leaderboards/"
date = "2025-10-03"
models/deepseek/deepseek-r1.toml
+51
View File
@@ -0,0 +1,51 @@
name = "DeepSeek-R1"
description = "Classic open reasoning model for transparent math, coding, and deliberate problem solving"
family = "deepseek-thinking"
release_date = "2025-01-20"
last_updated = "2025-05-29"
attachment = false
reasoning = true
temperature = true
tool_call = true
knowledge = "2024-07"
open_weights = true
[limit]
context = 128_000
output = 32_768
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/deepseek-ai/DeepSeek-R1"
[[benchmarks]]
name = "Aider Polyglot"
score = 56.9
metric = "percent correct"
source = "https://aider.chat/docs/leaderboards/"
date = "2025-01-20"
[[benchmarks]]
name = "Artificial Analysis Coding Index"
score = 15.9
metric = "index"
source = "https://openrouter.ai/deepseek/deepseek-r1/benchmarks"
date = "2026-03-11"
[[benchmarks]]
name = "SciCode"
score = 35.7
metric = "percent correct"
source = "https://openrouter.ai/deepseek/deepseek-r1/benchmarks"
date = "2026-03-11"
[[benchmarks]]
name = "Terminal-Bench Hard"
score = 6.1
metric = "success rate"
source = "https://openrouter.ai/deepseek/deepseek-r1/benchmarks"
date = "2026-03-11"
models/deepseek/deepseek-reasoner.toml
+30
View File
@@ -0,0 +1,30 @@
name = "DeepSeek Reasoner"
description = "DeepSeek reasoning model for multi-step analysis, math, coding, and tools"
family = "deepseek-thinking"
release_date = "2025-12-01"
last_updated = "2026-02-28"
attachment = true
reasoning = true
temperature = true
tool_call = true
knowledge = "2025-09"
open_weights = true
[limit]
context = 1_000_000
output = 384_000
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/deepseek-ai/DeepSeek-V3.2"
[[benchmarks]]
name = "Aider Polyglot"
score = 74.2
metric = "percent correct"
source = "https://aider.chat/docs/leaderboards/"
date = "2025-10-03"
models/deepseek/deepseek-v4-flash.toml
+30
View File
@@ -0,0 +1,30 @@
name = "DeepSeek V4 Flash"
description = "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work"
family = "deepseek-flash"
release_date = "2026-04-24"
last_updated = "2026-04-24"
attachment = false
reasoning = true
temperature = true
tool_call = true
structured_output = true
knowledge = "2025-05"
open_weights = true
[limit]
context = 1_000_000
output = 384_000
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash"
[[benchmarks]]
name = "SWE-Bench Verified"
score = 79
metric = "resolved"
source = "https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash"
models/deepseek/deepseek-v4-pro.toml
+64
View File
@@ -0,0 +1,64 @@
name = "DeepSeek V4 Pro"
description = "Open MoE flagship with million-token context for coding and long agent runs"
family = "deepseek-thinking"
release_date = "2026-04-24"
last_updated = "2026-04-24"
attachment = false
reasoning = true
temperature = true
tool_call = true
structured_output = true
knowledge = "2025-05"
open_weights = true
[limit]
context = 1_000_000
output = 384_000
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro"
[[benchmarks]]
name = "SWE-Bench Verified"
score = 80.6
metric = "resolved"
source = "https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro"
[[benchmarks]]
name = "Artificial Analysis Coding Agent Index"
score = 50.1
metric = "average pass@1"
harness = "Claude Code"
variant = "high"
source = "https://artificialanalysis.ai/agents/coding-agents"
[[benchmarks]]
name = "SWE-Atlas Codebase QnA"
score = 67.8
metric = "pass@1"
harness = "Claude Code"
variant = "high"
source = "https://artificialanalysis.ai/agents/coding-agents"
[[benchmarks]]
name = "SWE-Bench Pro"
score = 18
metric = "pass@1"
harness = "Claude Code"
variant = "high"
dataset = "hard-aa"
source = "https://artificialanalysis.ai/agents/coding-agents"
[[benchmarks]]
name = "Terminal-Bench"
score = 64.7
metric = "pass@1"
harness = "Claude Code"
variant = "high"
version = "2.1"
source = "https://artificialanalysis.ai/agents/coding-agents"
models/google/gemini-2.0-flash-lite.toml
+20
View File
@@ -0,0 +1,20 @@
name = "Gemini 2.0 Flash-Lite"
description = "Low-latency Gemini model for high-volume multimodal and agent workloads"
family = "gemini-flash-lite"
release_date = "2024-12-11"
last_updated = "2024-12-11"
attachment = true
reasoning = false
temperature = true
tool_call = true
structured_output = true
knowledge = "2024-06"
open_weights = false
[limit]
context = 1_048_576
output = 8_192
[modalities]
input = ["text", "image", "audio", "video", "pdf"]
output = ["text"]
models/google/gemini-2.0-flash.toml
+20
View File
@@ -0,0 +1,20 @@
name = "Gemini 2.0 Flash"
description = "Earlier Gemini Flash workhorse for responsive multimodal apps and tool use"
family = "gemini-flash"
release_date = "2024-12-11"
last_updated = "2024-12-11"
attachment = true
reasoning = false
temperature = true
tool_call = true
structured_output = true
knowledge = "2024-06"
open_weights = false
[limit]
context = 1_048_576
output = 8_192
[modalities]
input = ["text", "image", "audio", "video", "pdf"]
output = ["text"]
models/google/gemini-2.5-flash-image.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Nano Banana"
description = "Nano Banana image model for fast generation, edits, and character-consistent assets"
family = "gemini-flash"
release_date = "2025-08-26"
last_updated = "2025-08-26"
attachment = true
reasoning = true
temperature = true
tool_call = false
knowledge = "2024-06"
open_weights = false
[limit]
context = 32_768
output = 32_768
[modalities]
input = ["text", "image"]
output = ["text", "image"]
models/google/gemini-2.5-flash-lite.toml
+41
View File
@@ -0,0 +1,41 @@
name = "Gemini 2.5 Flash-Lite"
description = "Lean Gemini 2.5 lane for cheap multimodal traffic and quick agents"
family = "gemini-flash-lite"
release_date = "2025-06-17"
last_updated = "2025-06-17"
attachment = true
reasoning = true
temperature = true
tool_call = true
structured_output = true
knowledge = "2025-01"
open_weights = false
[limit]
context = 1_048_576
output = 65_536
[modalities]
input = ["text", "image", "audio", "video", "pdf"]
output = ["text"]
[[benchmarks]]
name = "Artificial Analysis Coding Index"
score = 9.5
metric = "index"
source = "https://openrouter.ai/google/gemini-2.5-flash-lite/benchmarks"
date = "2026-03-11"
[[benchmarks]]
name = "SciCode"
score = 19.3
metric = "percent correct"
source = "https://openrouter.ai/google/gemini-2.5-flash-lite/benchmarks"
date = "2026-03-11"
[[benchmarks]]
name = "Terminal-Bench Hard"
score = 4.5
metric = "success rate"
source = "https://openrouter.ai/google/gemini-2.5-flash-lite/benchmarks"
date = "2026-03-11"
models/google/gemini-2.5-flash-tts.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Gemini 2.5 Flash TTS"
description = "Speech generation model for controllable voice, narration, and audio delivery"
family = "gemini-flash"
release_date = "2025-09-30"
last_updated = "2025-12-10"
attachment = false
reasoning = false
temperature = true
tool_call = false
knowledge = "2025-01"
open_weights = false
[limit]
context = 32_768
output = 16_384
[modalities]
input = ["text"]
output = ["audio"]
models/google/gemini-2.5-flash.toml
+48
View File
@@ -0,0 +1,48 @@
name = "Gemini 2.5 Flash"
description = "Fast Gemini workhorse for multimodal apps where latency and price matter"
family = "gemini-flash"
release_date = "2025-06-17"
last_updated = "2025-06-17"
attachment = true
reasoning = true
temperature = true
tool_call = true
structured_output = true
knowledge = "2025-01"
open_weights = false
[limit]
context = 1_048_576
output = 65_536
[modalities]
input = ["text", "image", "audio", "video", "pdf"]
output = ["text"]
[[benchmarks]]
name = "Aider Polyglot"
score = 55.1
metric = "percent correct"
source = "https://aider.chat/docs/leaderboards/"
date = "2025-05-25"
[[benchmarks]]
name = "Artificial Analysis Coding Index"
score = 22.2
metric = "index"
source = "https://openrouter.ai/google/gemini-2.5-flash/benchmarks"
date = "2026-06-02"
[[benchmarks]]
name = "SciCode"
score = 39.4
metric = "percent correct"
source = "https://openrouter.ai/google/gemini-2.5-flash/benchmarks"
date = "2026-06-02"
[[benchmarks]]
name = "Terminal-Bench Hard"
score = 13.6
metric = "success rate"
source = "https://openrouter.ai/google/gemini-2.5-flash/benchmarks"
date = "2026-06-02"
models/google/gemini-2.5-pro-tts.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Gemini 2.5 Pro TTS"
description = "Speech generation model for controllable voice, narration, and audio delivery"
family = "gemini-pro"
release_date = "2025-09-30"
last_updated = "2025-12-10"
attachment = false
reasoning = false
temperature = false
tool_call = false
knowledge = "2025-01"
open_weights = false
[limit]
context = 32_768
output = 16_384
[modalities]
input = ["text"]
output = ["audio"]
models/google/gemini-2.5-pro.toml
+48
View File
@@ -0,0 +1,48 @@
name = "Gemini 2.5 Pro"
description = "Google's proven reasoning model for coding, math, and multimodal analysis"
family = "gemini-pro"
release_date = "2025-06-17"
last_updated = "2025-06-17"
attachment = true
reasoning = true
temperature = true
tool_call = true
structured_output = true
knowledge = "2025-01"
open_weights = false
[limit]
context = 1_048_576
output = 65_536
[modalities]
input = ["text", "image", "audio", "video", "pdf"]
output = ["text"]
[[benchmarks]]
name = "Aider Polyglot"
score = 83.1
metric = "percent correct"
source = "https://aider.chat/docs/leaderboards/"
date = "2025-06-06"
[[benchmarks]]
name = "Artificial Analysis Coding Index"
score = 32
metric = "index"
source = "https://openrouter.ai/google/gemini-2.5-pro/benchmarks"
date = "2026-06-02"
[[benchmarks]]
name = "SciCode"
score = 42.8
metric = "percent correct"
source = "https://openrouter.ai/google/gemini-2.5-pro/benchmarks"
date = "2026-06-02"
[[benchmarks]]
name = "Terminal-Bench Hard"
score = 26.5
metric = "success rate"
source = "https://openrouter.ai/google/gemini-2.5-pro/benchmarks"
date = "2026-06-02"
models/google/gemini-3-flash-preview.toml
+48
View File
@@ -0,0 +1,48 @@
name = "Gemini 3 Flash Preview"
description = "New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs"
family = "gemini-flash"
release_date = "2025-12-17"
last_updated = "2025-12-17"
attachment = true
reasoning = true
temperature = true
tool_call = true
structured_output = true
knowledge = "2025-01"
open_weights = false
[limit]
context = 1_048_576
output = 65_536
[modalities]
input = ["text", "image", "video", "audio", "pdf"]
output = ["text"]
[[benchmarks]]
name = "SWE-Bench Pro"
score = 34.63
metric = "resolve rate"
dataset = "public"
source = "https://labs.scale.com/leaderboard/swe_bench_pro_public"
[[benchmarks]]
name = "SWE-Atlas Codebase QnA"
score = 8.2
metric = "score"
harness = "Mini-SWE-Agent"
source = "https://labs.scale.com/leaderboard/sweatlas-qna"
[[benchmarks]]
name = "SWE-Atlas Refactoring"
score = 10
metric = "score"
harness = "Mini-SWE-Agent"
source = "https://labs.scale.com/leaderboard/sweatlas-refactoring"
[[benchmarks]]
name = "SWE-Atlas Test Writing"
score = 30.3
metric = "score"
harness = "Mini-SWE-Agent"
source = "https://labs.scale.com/leaderboard/sweatlas-tw"
models/google/gemini-3-pro-image-preview.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Nano Banana Pro"
description = "Nano Banana Pro for higher-fidelity image generation and design-heavy edits"
family = "gemini-pro"
release_date = "2025-11-20"
last_updated = "2025-11-20"
attachment = true
reasoning = true
temperature = true
tool_call = false
knowledge = "2025-01"
open_weights = false
[limit]
context = 65_536
output = 32_768
[modalities]
input = ["text", "image"]
output = ["text", "image"]
models/google/gemini-3-pro-image.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Nano Banana Pro"
description = "Nano Banana Pro for higher-fidelity image generation and design-heavy edits"
family = "gemini-pro"
release_date = "2026-05-28"
last_updated = "2026-05-28"
attachment = true
reasoning = true
temperature = true
tool_call = false
knowledge = "2025-01"
open_weights = false
[limit]
context = 65_536
output = 32_768
[modalities]
input = ["text", "image"]
output = ["text", "image"]
models/google/gemini-3-pro-preview.toml
+27
View File
@@ -0,0 +1,27 @@
name = "Gemini 3 Pro Preview"
description = "Preview Gemini flagship for complex reasoning, coding, and rich multimodal prompts"
family = "gemini-pro"
release_date = "2025-11-18"
last_updated = "2025-11-18"
attachment = true
reasoning = true
temperature = true
tool_call = true
structured_output = true
knowledge = "2025-01"
open_weights = false
[limit]
context = 1_048_576
output = 65_536
[modalities]
input = ["text", "image", "video", "audio", "pdf"]
output = ["text"]
[[benchmarks]]
name = "SWE-Bench Pro"
score = 43.3
metric = "resolve rate"
dataset = "public"
source = "https://labs.scale.com/leaderboard/swe_bench_pro_public"
models/google/gemini-3.1-flash-image-preview.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Nano Banana 2"
description = "Image model for prompt-driven generation, editing, and visual design workflows"
family = "gemini-flash"
release_date = "2026-02-26"
last_updated = "2026-02-26"
attachment = true
reasoning = true
temperature = true
tool_call = false
knowledge = "2025-01"
open_weights = false
[limit]
context = 65_536
output = 65_536
[modalities]
input = ["text", "image", "pdf"]
output = ["text", "image"]
models/google/gemini-3.1-flash-image.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Nano Banana 2"
description = "Image model for prompt-driven generation, editing, and visual design workflows"
family = "gemini-flash"
release_date = "2026-05-28"
last_updated = "2026-05-28"
attachment = true
reasoning = true
temperature = true
tool_call = false
knowledge = "2025-01"
open_weights = false
[limit]
context = 131_072
output = 32_768
[modalities]
input = ["text", "image", "video", "pdf"]
output = ["text", "image"]
models/google/gemini-3.1-flash-lite-preview.toml
+20
View File
@@ -0,0 +1,20 @@
name = "Gemini 3.1 Flash Lite Preview"
description = "Low-latency Gemini model for high-volume multimodal and agent workloads"
family = "gemini-flash-lite"
release_date = "2026-03-03"
last_updated = "2026-03-03"
attachment = true
reasoning = true
temperature = true
tool_call = true
structured_output = true
knowledge = "2025-01"
open_weights = false
[limit]
context = 1_048_576
output = 65_536
[modalities]
input = ["text", "image", "video", "audio", "pdf"]
output = ["text"]
models/google/gemini-3.1-flash-lite.toml
+20
View File
@@ -0,0 +1,20 @@
name = "Gemini 3.1 Flash Lite"
description = "Low-latency Gemini model for high-volume multimodal and agent workloads"
family = "gemini-flash-lite"
release_date = "2026-05-07"
last_updated = "2026-05-07"
attachment = true
reasoning = true
temperature = true
tool_call = true
structured_output = true
knowledge = "2025-01"
open_weights = false
[limit]
context = 1_048_576
output = 65_536
[modalities]
input = ["text", "image", "video", "audio", "pdf"]
output = ["text"]
models/google/gemini-3.1-pro-preview-customtools.toml
+20
View File
@@ -0,0 +1,20 @@
name = "Gemini 3.1 Pro Preview Custom Tools"
description = "Advanced Gemini model for complex reasoning, coding, and multimodal analysis"
family = "gemini-pro"
release_date = "2026-02-19"
last_updated = "2026-02-19"
attachment = true
reasoning = true
temperature = true
tool_call = true
structured_output = true
knowledge = "2025-01"
open_weights = false
[limit]
context = 1_048_576
output = 65_536
[modalities]
input = ["text", "image", "video", "audio", "pdf"]
output = ["text"]
models/google/gemini-3.1-pro-preview.toml
+157
View File
@@ -0,0 +1,157 @@
name = "Gemini 3.1 Pro Preview"
description = "Reasoning-first Gemini preview for agentic coding and complex problem solving"
family = "gemini-pro"
release_date = "2026-02-19"
last_updated = "2026-02-19"
attachment = true
reasoning = true
temperature = true
tool_call = true
structured_output = true
knowledge = "2025-01"
open_weights = false
[limit]
context = 1_048_576
output = 65_536
[modalities]
input = ["text", "image", "video", "audio", "pdf"]
output = ["text"]
[[benchmarks]]
name = "SWE-Bench Pro"
score = 54.2
metric = "resolve rate"
source = "https://www.anthropic.com/news/claude-opus-4-8"
date = "2026-05-28"
[[benchmarks]]
name = "Terminal-Bench"
score = 70.3
metric = "success rate"
harness = "Terminus-2"
version = "2.1"
source = "https://www.anthropic.com/news/claude-opus-4-8"
date = "2026-05-28"
[[benchmarks]]
name = "SWE-Bench Pro"
score = 46.1
metric = "resolve rate"
dataset = "public"
source = "https://labs.scale.com/leaderboard/swe_bench_pro_public"
[[benchmarks]]
name = "SWE-Atlas Codebase QnA"
score = 13.5
metric = "score"
harness = "Mini-SWE-Agent"
source = "https://labs.scale.com/leaderboard/sweatlas-qna"
[[benchmarks]]
name = "SWE-Atlas Refactoring"
score = 33.81
metric = "score"
harness = "Gemini CLI"
source = "https://labs.scale.com/leaderboard/sweatlas-refactoring"
[[benchmarks]]
name = "SWE-Atlas Test Writing"
score = 29.84
metric = "score"
harness = "Mini-SWE-Agent"
source = "https://labs.scale.com/leaderboard/sweatlas-tw"
[[benchmarks]]
name = "Artificial Analysis Coding Agent Index"
score = 43
metric = "average pass@1"
harness = "Gemini CLI"
variant = "high"
source = "https://artificialanalysis.ai/agents/coding-agents"
[[benchmarks]]
name = "SWE-Atlas Codebase QnA"
score = 45.6
metric = "pass@1"
harness = "Gemini CLI"
variant = "high"
source = "https://artificialanalysis.ai/agents/coding-agents"
[[benchmarks]]
name = "SWE-Bench Pro"
score = 15.1
metric = "pass@1"
harness = "Gemini CLI"
variant = "high"
dataset = "hard-aa"
source = "https://artificialanalysis.ai/agents/coding-agents"
[[benchmarks]]
name = "Terminal-Bench"
score = 68.3
metric = "pass@1"
harness = "Gemini CLI"
variant = "high"
version = "2.1"
source = "https://artificialanalysis.ai/agents/coding-agents"
[[benchmarks]]
name = "GPQA Diamond"
score = 94.3
metric = "accuracy"
source = "https://openai.com/index/introducing-gpt-5-5/"
date = "2026-04-23"
[[benchmarks]]
name = "Humanity's Last Exam"
score = 44.4
metric = "accuracy"
dataset = "full set, text + MM"
source = "https://deepmind.google/models/gemini/flash/"
date = "2026-05-19"
[[benchmarks]]
name = "ARC-AGI-2"
score = 77.1
metric = "accuracy"
source = "https://deepmind.google/models/gemini/flash/"
date = "2026-05-19"
[[benchmarks]]
name = "MMMU Pro"
score = 80.5
metric = "accuracy"
variant = "no tools"
source = "https://deepmind.google/models/gemini/flash/"
date = "2026-05-19"
[[benchmarks]]
name = "MCP Atlas"
score = 78.2
metric = "success rate"
source = "https://deepmind.google/models/gemini/flash/"
date = "2026-05-19"
[[benchmarks]]
name = "OSWorld-Verified"
score = 76.2
metric = "success rate"
source = "https://deepmind.google/models/gemini/flash/"
date = "2026-05-19"
[[benchmarks]]
name = "CharXiv Reasoning"
score = 83.3
metric = "accuracy"
variant = "no tools"
source = "https://deepmind.google/models/gemini/flash/"
date = "2026-05-19"
[[benchmarks]]
name = "GDPval-AA"
score = 1314
metric = "Elo"
source = "https://deepmind.google/models/gemini/flash/"
date = "2026-05-19"
models/google/gemini-3.5-flash.toml
+97
View File
@@ -0,0 +1,97 @@
name = "Gemini 3.5 Flash"
description = "Fast Gemini model balancing multimodal reasoning, tool use, and cost"
family = "gemini-flash"
release_date = "2026-05-19"
last_updated = "2026-05-19"
attachment = true
reasoning = true
temperature = true
tool_call = true
structured_output = true
knowledge = "2025-01"
open_weights = false
[limit]
context = 1_048_576
output = 65_536
[modalities]
input = ["text", "image", "video", "audio", "pdf"]
output = ["text"]
[[benchmarks]]
name = "Terminal-Bench"
score = 76.2
metric = "success rate"
harness = "Terminus-2"
version = "2.1"
source = "https://deepmind.google/models/gemini/flash/"
date = "2026-05-19"
[[benchmarks]]
name = "SWE-Bench Pro"
score = 55.1
metric = "resolve rate"
variant = "single attempt"
dataset = "public"
source = "https://deepmind.google/models/gemini/flash/"
date = "2026-05-19"
[[benchmarks]]
name = "MCP Atlas"
score = 83.6
metric = "success rate"
source = "https://deepmind.google/models/gemini/flash/"
date = "2026-05-19"
[[benchmarks]]
name = "Toolathlon"
score = 56.5
metric = "success rate"
source = "https://deepmind.google/models/gemini/flash/"
date = "2026-05-19"
[[benchmarks]]
name = "OSWorld-Verified"
score = 78.4
metric = "success rate"
source = "https://deepmind.google/models/gemini/flash/"
date = "2026-05-19"
[[benchmarks]]
name = "MMMU Pro"
score = 83.6
metric = "accuracy"
variant = "no tools"
source = "https://deepmind.google/models/gemini/flash/"
date = "2026-05-19"
[[benchmarks]]
name = "CharXiv Reasoning"
score = 84.2
metric = "accuracy"
variant = "no tools"
source = "https://deepmind.google/models/gemini/flash/"
date = "2026-05-19"
[[benchmarks]]
name = "Humanity's Last Exam"
score = 40.2
metric = "accuracy"
dataset = "full set, text + MM"
source = "https://deepmind.google/models/gemini/flash/"
date = "2026-05-19"
[[benchmarks]]
name = "ARC-AGI-2"
score = 72.1
metric = "accuracy"
source = "https://deepmind.google/models/gemini/flash/"
date = "2026-05-19"
[[benchmarks]]
name = "GDPval-AA"
score = 1656
metric = "Elo"
source = "https://deepmind.google/models/gemini/flash/"
date = "2026-05-19"
models/google/gemini-embedding-001.toml
+19
View File
@@ -0,0 +1,19 @@
name = "Gemini Embedding 001"
description = "Embedding model for semantic search, retrieval, clustering, and ranking pipelines"
family = "gemini"
release_date = "2025-05-20"
last_updated = "2025-05-20"
attachment = false
reasoning = false
temperature = false
tool_call = false
knowledge = "2025-05"
open_weights = false
[limit]
context = 2_048
output = 1
[modalities]
input = ["text"]
output = ["text"]
models/google/gemini-flash-latest.toml
+20
View File
@@ -0,0 +1,20 @@
name = "Gemini Flash Latest"
description = "Fast Gemini model balancing multimodal reasoning, tool use, and cost"
family = "gemini-flash"
release_date = "2025-09-25"
last_updated = "2025-09-25"
attachment = true
reasoning = true
temperature = true
tool_call = true
structured_output = true
knowledge = "2025-01"
open_weights = false
[limit]
context = 1_048_576
output = 65_536
[modalities]
input = ["text", "image", "audio", "video", "pdf"]
output = ["text"]
models/google/gemini-flash-lite-latest.toml
+20
View File
@@ -0,0 +1,20 @@
name = "Gemini Flash-Lite Latest"
description = "Low-latency Gemini model for high-volume multimodal and agent workloads"
family = "gemini-flash-lite"
release_date = "2025-09-25"
last_updated = "2025-09-25"
attachment = true
reasoning = true
temperature = true
tool_call = true
structured_output = true
knowledge = "2025-01"
open_weights = false
[limit]
context = 1_048_576
output = 65_536
[modalities]
input = ["text", "image", "audio", "video", "pdf"]
output = ["text"]
models/google/gemini-omni-flash-preview.toml
+24
View File
@@ -0,0 +1,24 @@
name = "Gemini Omni Flash Preview"
description = "Video generation and editing model for fast, conversational text- and image-to-video workflows"
family = "gemini"
release_date = "2026-06-30"
last_updated = "2026-06-30"
attachment = true
reasoning = true
tool_call = false
open_weights = false
[limit]
context = 1_048_576
output = 57_920
[modalities]
input = ["text", "image", "video"]
output = ["video"]
[[benchmarks]]
name = "LMArena Text-to-Video Arena"
score = 1527
metric = "Elo"
source = "https://venturebeat.com/technology/googles-gemini-omni-flash-hits-the-api-turning-enterprise-video-production-into-a-conversation"
date = "2026-06-30"
models/google/gemma-4-26b-a4b-it.toml
+23
View File
@@ -0,0 +1,23 @@
name = "Gemma 4 26B A4B IT"
description = "Open Gemma instruction model for efficient chat and self-hosted deployments"
family = "gemma"
release_date = "2026-04-02"
last_updated = "2026-04-02"
attachment = true
reasoning = true
temperature = true
tool_call = true
structured_output = true
open_weights = true
[limit]
context = 262_144
output = 32_768
[modalities]
input = ["text", "image"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/google/gemma-4-26B-A4B-it"
models/google/gemma-4-31b-it.toml
+23
View File
@@ -0,0 +1,23 @@
name = "Gemma 4 31B IT"
description = "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning"
family = "gemma"
release_date = "2026-04-02"
last_updated = "2026-04-02"
attachment = true
reasoning = true
temperature = true
tool_call = true
structured_output = true
open_weights = true
[limit]
context = 262_144
output = 32_768
[modalities]
input = ["text", "image"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/google/gemma-4-31B-it"
models/google/gemma-4-E2B-it.toml
+23
View File
@@ -0,0 +1,23 @@
name = "Gemma 4 E2B IT"
description = "Open Gemma instruction model for efficient chat and self-hosted deployments"
family = "gemma"
release_date = "2026-04-02"
last_updated = "2026-04-02"
attachment = true
reasoning = true
temperature = true
tool_call = true
structured_output = true
open_weights = true
[limit]
context = 131_072
output = 8_192
[modalities]
input = ["text", "image", "audio"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/google/gemma-4-E2B-it"
models/google/gemma-4-E4B-it.toml
+23
View File
@@ -0,0 +1,23 @@
name = "Gemma 4 E4B IT"
description = "Open Gemma instruction model for efficient chat and self-hosted deployments"
family = "gemma"
release_date = "2026-04-02"
last_updated = "2026-04-02"
attachment = true
reasoning = true
temperature = true
tool_call = true
structured_output = true
open_weights = true
[limit]
context = 131_072
output = 8_192
[modalities]
input = ["text", "image", "audio"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/google/gemma-4-E4B-it"
models/meituan/longcat-2.0.toml
+68
View File
@@ -0,0 +1,68 @@
name = "LongCat-2.0"
description = "Meituan LongCat-2.0, a reasoning model with tool calling and a 1M-token context window"
family = "longcat"
attachment = false
reasoning = true
temperature = true
tool_call = true
release_date = "2026-06-30"
last_updated = "2026-06-30"
open_weights = false
[limit]
context = 1_000_000
output = 131_072
[modalities]
input = ["text"]
output = ["text"]
[[benchmarks]]
name = "SWE-Bench Pro"
score = 59.5
metric = "resolve rate"
source = "https://github.com/meituan-longcat/longcat-2.0"
date = "2026-06-30"
[[benchmarks]]
name = "SWE-Bench Multilingual"
score = 77.3
metric = "resolve rate"
source = "https://github.com/meituan-longcat/longcat-2.0"
date = "2026-06-30"
[[benchmarks]]
name = "Terminal-Bench"
score = 70.8
metric = "success rate"
version = "2.1"
source = "https://github.com/meituan-longcat/longcat-2.0"
date = "2026-06-30"
[[benchmarks]]
name = "GPQA Diamond"
score = 88.9
metric = "accuracy"
source = "https://github.com/meituan-longcat/longcat-2.0"
date = "2026-06-30"
[[benchmarks]]
name = "BrowseComp"
score = 79.9
metric = "accuracy"
source = "https://github.com/meituan-longcat/longcat-2.0"
date = "2026-06-30"
[[benchmarks]]
name = "IFEval"
score = 90.0
metric = "accuracy"
source = "https://github.com/meituan-longcat/longcat-2.0"
date = "2026-06-30"
[[benchmarks]]
name = "FORTE"
score = 73.2
metric = "success rate"
source = "https://github.com/meituan-longcat/longcat-2.0"
date = "2026-06-30"
models/meta/llama-3.3-70b-instruct.toml
+44
View File
@@ -0,0 +1,44 @@
name = "Llama-3.3-70B-Instruct"
description = "Popular open Llama workhorse for multilingual chat, coding, and self-hosting"
family = "llama"
release_date = "2024-12-06"
last_updated = "2024-12-06"
attachment = true
reasoning = false
temperature = true
tool_call = true
knowledge = "2023-12"
open_weights = true
[limit]
context = 128_000
output = 4_096
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct"
[[benchmarks]]
name = "Artificial Analysis Coding Index"
score = 10.7
metric = "index"
source = "https://openrouter.ai/meta-llama/llama-3.3-70b-instruct/benchmarks"
date = "2026-03-11"
[[benchmarks]]
name = "SciCode"
score = 26
metric = "percent correct"
source = "https://openrouter.ai/meta-llama/llama-3.3-70b-instruct/benchmarks"
date = "2026-03-11"
[[benchmarks]]
name = "Terminal-Bench Hard"
score = 3
metric = "success rate"
source = "https://openrouter.ai/meta-llama/llama-3.3-70b-instruct/benchmarks"
date = "2026-03-11"
models/meta/llama-4-maverick-17b-instruct.toml
+37
View File
@@ -0,0 +1,37 @@
name = "Llama 4 Maverick 17B Instruct"
description = "Open multimodal Llama for strong reasoning with efficient everyday serving"
family = "llama"
release_date = "2025-04-05"
last_updated = "2025-04-05"
attachment = true
reasoning = false
temperature = true
tool_call = true
knowledge = "2024-08"
open_weights = true
[limit]
context = 1_000_000
output = 16_384
[modalities]
input = ["text", "image"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/meta-llama/Llama-4-Maverick-17B-128E-Instruct"
[[benchmarks]]
name = "Aider Polyglot"
score = 15.6
metric = "percent correct"
source = "https://aider.chat/docs/leaderboards/"
date = "2025-04-06"
[[benchmarks]]
name = "SWE-Bench Pro"
score = 5.24
metric = "resolve rate"
dataset = "public"
source = "https://labs.scale.com/leaderboard/swe_bench_pro_public"
models/meta/llama-4-scout-17b-instruct.toml
+23
View File
@@ -0,0 +1,23 @@
name = "Llama 4 Scout 17B Instruct"
description = "Open Llama with long-context vision for efficient multimodal agents"
family = "llama"
release_date = "2025-04-05"
last_updated = "2025-04-05"
attachment = true
reasoning = false
temperature = true
tool_call = true
knowledge = "2024-08"
open_weights = true
[limit]
context = 3_500_000
output = 16_384
[modalities]
input = ["text", "image"]
output = ["text"]
[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/meta-llama/Llama-4-Scout-17B-16E-Instruct"

Some files were not shown because too many files have changed in this diff Show More