Gemma 31B Sft 250 is an open-source language model by harisarang. Features: 31b LLM, VRAM: 62.5GB, LLM Explorer Score: 0.29.
Model Card on HF 🤗: https://huggingface.co/harisarang/gemma-31b-sft-250
Gemma 31B Sft 250 Benchmarks
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
What is the LLM Explorer Rank (Score)
Gemma 31B Sft 250 Parameters and Internals
| LLM Name | Gemma 31B Sft 250 |
| Repository 🤗 | https://huggingface.co/harisarang/gemma-31b-sft-250 |
| Model Size | 31b |
| Required VRAM | 62.5 GB |
| Updated | 2026-04-29 |
| Maintainer | harisarang |
| Model Type | gemma4 |
| Model Files | |
| Model Architecture | Gemma4ForConditionalGeneration |
| Transformers Version | 5.5.4 |
| Tokenizer Class | GemmaTokenizer |
| Padding Token | <pad> |
Best Alternatives to Gemma 31B Sft 250
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Gemma 4 31B IT NVFP4 | 0K / 32.6 GB | 1611755 | 425 |
| Gemma 4 31B It | 0K / 62.6 GB | 46845 | 13 |
| Gemma 4 31B It Bf16 | 0K / 62.6 GB | 7225 | 4 |
| Gemopus 4 31B It | 0K / 62.6 GB | 576 | 7 |
| Gemma4 GarnetV2 31B | 0K / 62.6 GB | 325 | 3 |
| Gemma4 Garnet 31B | 0K / 62.6 GB | 497 | 9 |
| Gemma 4 31B JANG 4M CRACK | 0K / 22.6 GB | 1089 | 2 |
| G4 31B Musica V1 | 0K / 62.5 GB | 0 | 5 |
| ...RD HERETIC UNCENSORED Thinking | 0K / 62.5 GB | 5899 | 13 |
| Gemma 4 31B It Mlx 4bit | 0K / 18.5 GB | 2329 | 4 |
Note: green Score (e.g. "73.2") means that the model is better than harisarang/gemma-31b-sft-250 .
Rank the Gemma 31B Sft 250 Capabilities
🆘 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! 🌟
What open-source LLMs or SLMs are you in search of? 53310 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a