eeffa4c065## Summary - Add web search tool calling for GGUF models (Search toggle, DuckDuckGo via ddgs) - Add KV cache dtype dropdown (f16/bf16/q8_0/q5_1/q4_1) in Chat Settings - Fix Qwen3/3.5 inference defaults per official docs (thinking on/off params) - Enable reasoning by default for Qwen3.5 4B and 9B - Replace "Generating" toast with inline spinner - Fix stop button via asyncio.to_thread (event loop no longer blocked) - Fix CUDA 12 compat lib paths for llama-server on CUDA 13 systems - Fix auto-load model name not appearing in selector - Training progress messages + dataset_num_proc fix Integrated PRs: - #4327 (imagineer99): BETA badge alignment (already in tree) - #4340 (Manan Shah): prioritize training models in model selection - #4344 (Roland Tannous): setup.sh macOS python version compatibility - #4345 (Manan Shah): revamp model+dataset checking logic
381 lines
8 KiB
JSON
381 lines
8 KiB
JSON
{
|
|
"_comment": "Per-model-family inference parameter defaults. Sources: (1) Ollama params blobs, (2) Existing Unsloth Studio YAML configs. Patterns ordered longest-match-first.",
|
|
"families": {
|
|
"qwen3.5": {
|
|
"temperature": 0.7,
|
|
"top_p": 0.8,
|
|
"top_k": 20,
|
|
"min_p": 0.0,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"qwen3-coder": {
|
|
"temperature": 0.7,
|
|
"top_p": 0.8,
|
|
"top_k": 20,
|
|
"min_p": 0.0,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"qwen3-next": {
|
|
"temperature": 0.7,
|
|
"top_p": 0.8,
|
|
"top_k": 20,
|
|
"min_p": 0.0,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"qwen3-vl": {
|
|
"temperature": 0.7,
|
|
"top_p": 0.8,
|
|
"top_k": 20,
|
|
"min_p": 0.0,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"qwen3": {
|
|
"temperature": 0.6,
|
|
"top_p": 0.95,
|
|
"top_k": 20,
|
|
"min_p": 0.0,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"qwen2.5-coder": {
|
|
"temperature": 1.5,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.1,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"qwen2.5-vl": {
|
|
"temperature": 1.5,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.1,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"qwen2.5-omni": {
|
|
"temperature": 0.7,
|
|
"top_p": 0.8,
|
|
"top_k": 20,
|
|
"min_p": 0.0,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"qwen2.5-math": {
|
|
"temperature": 0.7,
|
|
"top_p": 0.8,
|
|
"top_k": 20,
|
|
"min_p": 0.0,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"qwen2.5": {
|
|
"temperature": 0.7,
|
|
"top_p": 0.8,
|
|
"top_k": 20,
|
|
"min_p": 0.0,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"qwen2-vl": {
|
|
"temperature": 1.5,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.1,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"qwen2": {
|
|
"temperature": 0.7,
|
|
"top_p": 0.8,
|
|
"top_k": 20,
|
|
"min_p": 0.0,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"qwq": {
|
|
"temperature": 0.6,
|
|
"top_p": 0.95,
|
|
"top_k": 40,
|
|
"min_p": 0.0,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"gemma-3n": {
|
|
"temperature": 1.0,
|
|
"top_p": 0.95,
|
|
"top_k": 64,
|
|
"min_p": 0.0,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"gemma-3": {
|
|
"temperature": 1.0,
|
|
"top_p": 0.95,
|
|
"top_k": 64,
|
|
"min_p": 0.0,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"medgemma": {
|
|
"temperature": 1.0,
|
|
"top_p": 0.95,
|
|
"top_k": 64,
|
|
"min_p": 0.0,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"gemma-2": {
|
|
"temperature": 1.0,
|
|
"top_p": 0.95,
|
|
"top_k": 64,
|
|
"min_p": 0.0,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"llama-4": {
|
|
"temperature": 1.0,
|
|
"top_p": 0.9,
|
|
"top_k": -1,
|
|
"min_p": 0.01,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"llama-3.3": {
|
|
"temperature": 1.5,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.1,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"llama-3.2": {
|
|
"temperature": 1.5,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.1,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"llama-3.1": {
|
|
"temperature": 1.5,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.1,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"llama-3": {
|
|
"temperature": 1.5,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.1,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"phi-4": {
|
|
"temperature": 0.8,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.0,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"phi-3": {
|
|
"temperature": 0.7,
|
|
"top_p": 0.9,
|
|
"top_k": -1,
|
|
"min_p": 0.01,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"mistral-nemo": {
|
|
"temperature": 0.7,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.01,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"mistral-small": {
|
|
"temperature": 0.15,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.01,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"mistral-large": {
|
|
"temperature": 0.7,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.01,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"magistral": {
|
|
"temperature": 0.7,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.01,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"ministral": {
|
|
"temperature": 0.15,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.01,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"devstral": {
|
|
"temperature": 0.7,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.01,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"pixtral": {
|
|
"temperature": 1.5,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.1,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"deepseek-r1": {
|
|
"temperature": 0.6,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.01,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"deepseek-v3": {
|
|
"temperature": 0.6,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.01,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"deepseek-ocr": {
|
|
"temperature": 0.0,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.01,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"glm-5": {
|
|
"temperature": 1.0,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.01,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"glm-4": {
|
|
"temperature": 1.0,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.01,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"nemotron": {
|
|
"temperature": 1.0,
|
|
"top_p": 1.0,
|
|
"top_k": -1,
|
|
"min_p": 0.01,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"minimax-m2.5": {
|
|
"temperature": 1.0,
|
|
"top_p": 0.95,
|
|
"top_k": 40,
|
|
"min_p": 0.01,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"minimax": {
|
|
"temperature": 1.0,
|
|
"top_p": 0.95,
|
|
"top_k": 40,
|
|
"min_p": 0.01,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"gpt-oss": {
|
|
"temperature": 1.0,
|
|
"top_p": 1.0,
|
|
"top_k": 0,
|
|
"min_p": 0.01,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"granite-4": {
|
|
"temperature": 0.0,
|
|
"top_p": 1.0,
|
|
"top_k": 0,
|
|
"min_p": 0.01,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"kimi-k2": {
|
|
"temperature": 0.6,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.01,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"kimi": {
|
|
"temperature": 0.6,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.01,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"lfm2": {
|
|
"temperature": 0.1,
|
|
"top_p": 0.1,
|
|
"top_k": 50,
|
|
"min_p": 0.15,
|
|
"repetition_penalty": 1.05
|
|
},
|
|
"smollm": {
|
|
"temperature": 0.7,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.01,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"olmo": {
|
|
"temperature": 0.7,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.01,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"falcon": {
|
|
"temperature": 0.7,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.01,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"ernie": {
|
|
"temperature": 0.7,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.01,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"seed": {
|
|
"temperature": 0.7,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.01,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"grok": {
|
|
"temperature": 1.0,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.01,
|
|
"repetition_penalty": 1.0
|
|
},
|
|
"mimo": {
|
|
"temperature": 0.7,
|
|
"top_p": 0.95,
|
|
"top_k": -1,
|
|
"min_p": 0.01,
|
|
"repetition_penalty": 1.0
|
|
}
|
|
},
|
|
"patterns": [
|
|
"qwen3.5",
|
|
"qwen3-coder", "qwen3-next", "qwen3-vl", "qwen3",
|
|
"qwen2.5-coder", "qwen2.5-vl", "qwen2.5-omni", "qwen2.5-math", "qwen2.5",
|
|
"qwen2-vl", "qwen2",
|
|
"qwq",
|
|
"gemma-3n", "gemma-3", "medgemma", "gemma-2",
|
|
"llama-4", "llama-3.3", "llama-3.2", "llama-3.1", "llama-3",
|
|
"phi-4", "phi-3",
|
|
"mistral-nemo", "mistral-small", "mistral-large", "magistral", "ministral",
|
|
"devstral", "pixtral",
|
|
"deepseek-r1", "deepseek-v3", "deepseek-ocr",
|
|
"glm-5", "glm-4",
|
|
"nemotron",
|
|
"minimax-m2.5", "minimax",
|
|
"gpt-oss", "granite-4",
|
|
"kimi-k2", "kimi",
|
|
"lfm2", "smollm", "olmo", "falcon", "ernie", "seed", "grok", "mimo"
|
|
]
|
|
}
|