| Model Type | 
 | |||||||||
| Use Cases | 
 | |||||||||
| Additional Notes | 
 | 
| LLM Name | Codegemma 7B It Bnb 4bit | 
| Repository ๐ค | https://huggingface.co/unsloth/codegemma-7b-it-bnb-4bit | 
| Model Size | 7b | 
| Required VRAM | 5.6 GB | 
| Updated | 2025-09-23 | 
| Maintainer | unsloth | 
| Model Type | gemma | 
| Model Files | |
| Supported Languages | en | 
| Quantization Type | 4bit | 
| Model Architecture | GemmaForCausalLM | 
| License | apache-2.0 | 
| Context Length | 8192 | 
| Model Max Length | 8192 | 
| Transformers Version | 4.44.2 | 
| Tokenizer Class | GemmaTokenizer | 
| Padding Token | <pad> | 
| Vocabulary Size | 256000 | 
| Torch Data Type | bfloat16 | 
| Best Alternatives | Context / RAM | Downloads | Likes | 
|---|---|---|---|
| ...t Cleaner Gemma 32k Merged 16b | 31K / 17.1 GB | 5 | 0 | 
| ...ruct Malayalam Model Vllm 4bit | 16K / 5.6 GB | 5 | 0 | 
| ... Codegemma 2 7B It Alpaca V1.2 | 8K / 17.1 GB | 5 | 0 | 
| Codegemma 7B It 8bit | 8K / 9.5 GB | 1147 | 5 | 
| Fbt Gemma 7B | 8K / 17.1 GB | 5 | 0 | 
| Codegemma 7B It 4bit | 8K / 5.9 GB | 38 | 1 | 
| Codegemma 7B 4bit | 8K / 5.9 GB | 17 | 0 | 
| Gemma 1.1 7B It 4bit | 8K / 5.9 GB | 18 | 4 | 
| Gemma 1.1 7B It 4bit | 8K / 5.9 GB | 6 | 1 | 
| Swahili Gemma | 8K / 17.1 GB | 9 | 1 | 
๐ Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! ๐