chore: import upstream snapshot with attribution

2026-07-13 12:28:55 +08:00
commit db42b91b75
6397 changed files with 146012 additions and 0 deletions
@@ -0,0 +1,25 @@
+name = "Gemini 2.0 Flash-Lite"
+description = "Legacy model retained for compatibility with older integrations"
+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
+status = "deprecated"
+
+[cost]
+input = 0.075
+output = 0.3
+
+[limit]
+context = 1_048_576
+output = 8_192
+
+[modalities]
+input = ["text", "image", "audio", "video", "pdf"]
+output = ["text"]
@@ -0,0 +1,26 @@
+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
+knowledge = "2024-06"
+tool_call = true
+structured_output = true
+open_weights = false
+status = "deprecated"
+
+[cost]
+input = 0.10
+output = 0.40
+cache_read = 0.025
+
+[limit]
+context = 1_048_576
+output = 8_192
+
+[modalities]
+input = ["text", "image", "audio", "video", "pdf"]
+output = ["text"]
@@ -0,0 +1,7 @@
+base_model = "google/gemini-2.5-flash-image"
+reasoning_options = []
+
+[cost]
+input = 0.3
+output = 30
+cache_read = 0.075
@@ -0,0 +1,34 @@
+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
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "budget_tokens"
+min = 512
+max = 24_576
+
+[cost]
+input = 0.1
+output = 0.4
+cache_read = 0.01
+input_audio = 0.3
+
+[limit]
+context = 1_048_576
+output = 65_536
+
+[modalities]
+input = ["text", "image", "audio", "video", "pdf"]
+output = ["text"]
@@ -0,0 +1,23 @@
+name = "Gemini 2.5 Flash Preview TTS"
+description = "Speech generation model for controllable voice, narration, and audio delivery"
+family = "gemini-flash"
+release_date = "2025-05-01"
+last_updated = "2025-05-01"
+attachment = false
+reasoning = false
+temperature = true
+tool_call = false
+knowledge = "2025-01"
+open_weights = false
+
+[cost]
+input = 0.5
+output = 10
+
+[limit]
+context = 8_192
+output = 16_384
+
+[modalities]
+input = ["text"]
+output = ["audio"]
@@ -0,0 +1,15 @@
+base_model = "google/gemini-2.5-flash"
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "budget_tokens"
+min = 0
+max = 24_576
+
+[cost]
+input = 0.30
+output = 2.50
+cache_read = 0.03
+input_audio = 1.00
@@ -0,0 +1,23 @@
+name = "Gemini 2.5 Pro Preview TTS"
+description = "Speech generation model for controllable voice, narration, and audio delivery"
+family = "gemini-flash"
+release_date = "2025-05-01"
+last_updated = "2025-05-01"
+attachment = false
+reasoning = false
+temperature = true
+tool_call = false
+knowledge = "2025-01"
+open_weights = false
+
+[cost]
+input = 1
+output = 20
+
+[limit]
+context = 8_192
+output = 16_384
+
+[modalities]
+input = ["text"]
+output = ["audio"]
@@ -0,0 +1,17 @@
+base_model = "google/gemini-2.5-pro"
+
+[[reasoning_options]]
+type = "budget_tokens"
+min = 128
+max = 32_768
+
+[cost]
+input = 1.25
+output = 10.00
+cache_read = 0.125
+
+[[cost.tiers]]
+tier = { size = 200_000 }
+input = 2.50
+output = 15.00
+cache_read = 0.25
@@ -0,0 +1,30 @@
+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
+
+[[reasoning_options]]
+type = "effort"
+values = ["minimal", "low", "medium", "high"]
+
+[cost]
+input = 0.50
+output = 3.00
+cache_read = 0.05
+input_audio = 1.00
+
+[limit]
+context = 1048576
+output = 65536
+
+[modalities]
+input = ["text", "image", "video", "audio", "pdf"]
+output = ["text"]
@@ -0,0 +1,9 @@
+base_model = "google/gemini-3-pro-image-preview"
+reasoning_options = []
+
+[cost]
+input = 2
+output = 120
+
+[limit]
+context = 131_072
@@ -0,0 +1,36 @@
+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
+status = "deprecated"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "high"]
+
+[cost]
+input = 2
+output = 12
+cache_read = 0.2
+
+[[cost.tiers]]
+tier = { size = 200_000 }
+input = 4
+output = 18
+cache_read = 0.4
+
+[limit]
+context = 1_048_576
+output = 65_536
+
+[modalities]
+input = ["text", "image", "video", "audio", "pdf"]
+output = ["text"]
@@ -0,0 +1,27 @@
+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
+
+[[reasoning_options]]
+type = "effort"
+values = ["minimal", "high"]
+
+[cost]
+input = 0.5
+output = 60
+
+[limit]
+context = 65_536
+output = 65_536
+
+[modalities]
+input = ["text", "image", "pdf"]
+output = ["text", "image"]
@@ -0,0 +1,31 @@
+name = "Gemini 3.1 Flash Lite Preview"
+description = "Legacy model retained for compatibility with older integrations"
+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
+status = "deprecated"
+
+[[reasoning_options]]
+type = "effort"
+values = ["minimal", "low", "medium", "high"]
+
+[cost]
+input = 0.25
+output = 1.50
+cache_read = 0.025
+input_audio = 0.50
+
+[limit]
+context = 1048576
+output = 65536
+
+[modalities]
+input = ["text", "image", "video", "audio", "pdf"]
+output = ["text"]
@@ -0,0 +1,30 @@
+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
+
+[[reasoning_options]]
+type = "effort"
+values = ["minimal", "low", "medium", "high"]
+
+[cost]
+input = 0.25
+output = 1.50
+cache_read = 0.025
+input_audio = 0.50
+
+[limit]
+context = 1_048_576
+output = 65_536
+
+[modalities]
+input = ["text", "image", "video", "audio", "pdf"]
+output = ["text"]
@@ -0,0 +1,35 @@
+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
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high"]
+
+[cost]
+input = 2.00
+output = 12.00
+cache_read = 0.20
+
+[[cost.tiers]]
+tier = { size = 200_000 }
+input = 4.00
+output = 18.00
+cache_read = 0.40
+
+[limit]
+context = 1_048_576
+output = 65_536
+
+[modalities]
+input = ["text", "image", "video", "audio", "pdf"]
+output = ["text"]
@@ -0,0 +1,35 @@
+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
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high"]
+
+[cost]
+input = 2.00
+output = 12.00
+cache_read = 0.20
+
+[[cost.tiers]]
+tier = { size = 200_000 }
+input = 4.00
+output = 18.00
+cache_read = 0.40
+
+[limit]
+context = 1_048_576
+output = 65_536
+
+[modalities]
+input = ["text", "image", "video", "audio", "pdf"]
+output = ["text"]
@@ -0,0 +1,30 @@
+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
+
+[[reasoning_options]]
+type = "effort"
+values = ["minimal", "low", "medium", "high"]
+
+[cost]
+input = 1.50
+output = 9.00
+cache_read = 0.15
+input_audio = 1.50
+
+[limit]
+context = 1_048_576
+output = 65_536
+
+[modalities]
+input = ["text", "image", "video", "audio", "pdf"]
+output = ["text"]
@@ -0,0 +1,23 @@
+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
+
+[cost]
+input = 0.15
+output = 0
+
+[limit]
+context = 2_048
+output = 1
+
+[modalities]
+input = ["text"]
+output = ["text"]
@@ -0,0 +1,34 @@
+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
+knowledge = "2025-01"
+tool_call = true
+structured_output = true
+open_weights = false
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "budget_tokens"
+min = 0
+max = 24_576
+
+[cost]
+input = 0.30
+output = 2.50
+cache_read = 0.075
+input_audio = 1.00
+
+[limit]
+context = 1_048_576
+output = 65_536
+
+[modalities]
+input = ["text", "image", "audio", "video", "pdf"]
+output = ["text"]
@@ -0,0 +1,33 @@
+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
+knowledge = "2025-01"
+tool_call = true
+structured_output = true
+open_weights = false
+
+[[reasoning_options]]
+type = "toggle"
+
+[[reasoning_options]]
+type = "budget_tokens"
+min = 512
+max = 24_576
+
+[cost]
+input = 0.10
+output = 0.40
+cache_read = 0.025
+
+[limit]
+context = 1_048_576
+output = 65_536
+
+[modalities]
+input = ["text", "image", "audio", "video", "pdf"]
+output = ["text"]
@@ -0,0 +1,11 @@
+base_model = "google/gemini-omni-flash-preview"
+reasoning_options = []
+temperature = true
+
+[cost]
+input = 1.5
+output = 17.5
+
+[limit]
+context = 131_072
+output = 65_536
@@ -0,0 +1,20 @@
+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
+reasoning_options = [{ type = "toggle" }]
+
+[limit]
+context = 262_144
+output = 32_768
+
+[modalities]
+input = ["text", "image"]
+output = ["text"]
@@ -0,0 +1,20 @@
+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
+reasoning_options = [{ type = "toggle" }]
+
+[limit]
+context = 262_144
+output = 32_768
+
+[modalities]
+input = ["text", "image"]
+output = ["text"]