wehub-resource-sync
65 lines
1.4 KiB
TOML
65 lines
1.4 KiB
TOML
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"
|