====== Elenco dei modelli di linguaggio ======

^ Nome del modello ^ File GGUF ^ Parametri ^
|Qwen3.8-27B-UD-Q3_K_XL-96K|https://huggingface.co/unsloth/Qwen3.8-27B-GGUF|top-p=0.95, top-k=20, min-p=0.0, c=96000, temp=1.0, spec-type=draft-mtp, spec-draft-n-max=2, no-mmproj-offload=true|
|Qwen3.6-35B-A3B-UD-Q3_K_XL-T_0_6-128K|https://huggingface.co/unsloth/Qwen3.6-35B-A3B-MTP-GGUF|c=131172, temp=0.6, fa=true, gpu-layers=999, spec-type=draft-mtp, spec-draft-n-max=2|
|Qwen3.6-35B-A3B-UD-Q3_K_XL-T_0_6-256K|https://huggingface.co/unsloth/Qwen3.6-35B-A3B-MTP-GGUF|c=262144, temp=0.6, fa=true, gpu-layers=999, spec-type=draft-mtp, spec-draft-n-max=2|
|Qwen3.6-35B-A3B-UD-Q3_K_XL-128K|https://huggingface.co/unsloth/Qwen3.6-35B-A3B-MTP-GGUF|c=131172, fa=true, gpu-layers=999, spec-type=draft-mtp, spec-draft-n-max=2|
|Qwen3.6-35B-A3B-UD-Q3_K_XL-256K|https://huggingface.co/unsloth/Qwen3.6-35B-A3B-MTP-GGUF|c=262144, fa=true, gpu-layers=999, spec-type=draft-mtp, spec-draft-n-max=2|
|gemma-4-E2B-it-qat-UD-Q4_K_XL-32K|https://huggingface.co/unsloth/gemma-4-E2B-it-qat-GGUF|c=32000, gpu-layers=999, top-p=0.95, top-k=64|
|gemma-4-26B-A4B-it-qat-128K|https://huggingface.co/unsloth/gemma-4-26B-A4B-it-qat-GGUF|c=131172, gpu-layers=999, top-p=0.95, top-k=64|

----

===== OpenCode config (example) =====
<code jsonc>
{
    "$schema": "https://opencode.ai/config.json",
    "model": "Qwen3.8-27B-UD-Q3_K_XL",
    "provider": {
        "llama-cpp": {
            "npm": "@ai-sdk/openai-compatible",
            "options": {
                "baseURL": "https://llm.hq.mittelab.org/v1"
            },
            "models": {
                "Qwen3.8-27B-UD-Q3_K_XL": {
                    "tools": true,
                    "modalities": { "input": ["text"] },
                    "limit": { "context": 65535, "output": 8192 }
                }
            }
        }
    }
}

</code>

===== Pi config (example) =====

*~/.pi/agent/settings.json*
<code json>
{
    "defaultProvider": "llama.cpp",
    "defaultModel": "Qwen3.8-27B-UD-Q3_K_XL"
}

</code>

*~/.pi/agent/models.json*
<code json>
{
    "providers": {
        "llama.cpp": {
            "baseUrl": "https://llm.hq.mittelab.org",
            "api": "openai-completions",
            "apiKey": "llama.cpp",
            "models": [
                { "id": "Qwen3.8-27B-UD-Q3_K_XL", "contextWindow": 65535 }
            ]
        }
    }
}

</code>

*~/.pi/agent/auth.json*
<code json>
{
    "llama.cpp": {
        "type": "api_key",
        "env": {
            "LLAMA_BASE_URL": "https://llm.hq.mittelab.org"
        }
    }
}

</code>

----

**Nota**: Questa pagina è stata generata automaticamente, non modificare!