| --- |
| title: Gemma 4 26B Coding API |
| emoji: π |
| colorFrom: indigo |
| colorTo: green |
| sdk: docker |
| pinned: false |
| app_port: 7860 |
| --- |
| |
| # Gemma 4 26B A4B β Coding API |
|
|
| Gemma 4 26B A4B (MoE) served as an Anthropic + OpenAI compatible API. |
| Model: `unsloth/gemma-4-26B-A4B-it-GGUF` Β· `UD-IQ3_XXS` (11.2 GB) |
| Params: temp=0.3, top_p=0.9, min_p=0.1, top_k=20 (coding-tuned) |
| |
| ## Claude Code setup |
| |
| ```bash |
| export ANTHROPIC_BASE_URL=https://YOUR_USERNAME-gemma-4-26b-coding-api.hf.space |
| export ANTHROPIC_API_KEY=gemma4-local |
| claude --model gemma-4-26b |
| ``` |
| |
| ## Environment variables (HF Space β Settings β Variables) |
| |
| | Variable | Default | Description | |
| |----------|---------|-------------| |
| | `SPACE_URL` | `` | Your space URL β enables self-ping | |
| | `MODEL_REPO` | `unsloth/gemma-4-26B-A4B-it-GGUF` | HF repo | |
| | `MODEL_FILE` | `gemma-4-26B-A4B-it-UD-IQ3_XXS.gguf` | GGUF filename | |
| | `N_CTX` | `4096` | Context window | |
| | `N_THREADS` | `2` | CPU threads | |
| | `DEFAULT_TEMP` | `0.3` | Temperature | |
| | `DEFAULT_MIN_P` | `0.1` | Min-p (key for coding accuracy) | |
| | `DEFAULT_TOP_K` | `20` | Top-k | |