GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
ressl
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
ressl
Model Tree
google/gemma-4-31B-it
Input Modalities
Output Modalities
Supported Functionality
| Repository | Format | Runs on |
|---|---|---|
| ressl/gemma-4-31B-it-uncensored | Transformers BF16, multimodal | transformers, vLLM, SGLang |
| ressl/gemma-4-31B-it-uncensored-NVFP4 | NVIDIA NVFP4, multimodal | vLLM, SGLang on Blackwell |
| ressl/gemma-4-31B-it-uncensored-GGUF | GGUF q8_0 to q2_k, text only | llama.cpp, Ollama, LM Studio |
| ressl/gemma-4-31B-it-uncensored-MLX-bf16 | MLX BF16, multimodal | mlx-vlm on Apple silicon |
| ressl/gemma-4-31B-it-uncensored-MLX-8bit | MLX 8-bit, multimodal | mlx-vlm on Apple silicon |
| ressl/gemma-4-31B-it-uncensored-MLX-6bit | MLX 6-bit, multimodal | mlx-vlm on Apple silicon |
| ressl/gemma-4-31B-it-uncensored-MLX-5bit | MLX 5-bit, multimodal | mlx-vlm on Apple silicon |
| ressl/gemma-4-31B-it-uncensored-MLX-4bit | MLX 4-bit, multimodal | mlx-vlm on Apple silicon |
| Base model | google/gemma-4-31B-it (30.7B dense, multimodal) |
| Type | uncensored (abliterated) build |
| Effective refusals | 0 / 686 hard-refusal across 4 datasets (base 99/100) |
| KL divergence vs base | 0.1234 (low, capabilities preserved) |
| Coherence | intact, math, multilingual and factual answers correct |
| Size | 59 GB (BF16) |
| Hardware | 2× NVIDIA RTX PRO 6000 Blackwell 96 GB (SM120), driver 610 |
| Format | Engine | Hard refusals |
|---|---|---|
| BF16 | vLLM · SGLang · transformers | 0/100 |
| NVFP4 | vLLM · SGLang | 0/100 |
| GGUF f16…q2_k | llama.cpp (CUDA, ~75 tok/s) | 0-1/100 |
Generalization tested across 686 prompts from 4 independent datasets, 0 effective refusals everywhere:
| Dataset | Prompts | Effective refusals |
|---|---|---|
| JailbreakBench | 100 | 0/100 |
| tulu-harmbench | 320 | 0/320 |
| NousResearch/RefusalDataset | 166 | 0/166 |
| mlabonne/harmful_behaviors | 100 | 0/100 |
| Total | 686 | 0/686 (0.0%) |
A naive keyword detector flags 363/686 (52.9%), every one is a ***Disclaimer:**-prefixed
compliant answer, not a refusal.
bash
vllm serve ressl/gemma-4-31B-it-uncensored --max-model-len 8192 --trust-remote-code
bash
python -m sglang.launch_server --model-path ressl/gemma-4-31B-it-uncensored \--attention-backend triton --trust-remote-code
gemma-4 requires --attention-backend triton (it rejects flashinfer). Tested with sglang 0.5.14 on Python 3.12.
python
from transformers import AutoModelForImageTextToText, AutoTokenizerimport torchtok = AutoTokenizer.from_pretrained("ressl/gemma-4-31B-it-uncensored")m = AutoModelForImageTextToText.from_pretrained("ressl/gemma-4-31B-it-uncensored", dtype=torch.bfloat16, device_map="cuda")
***Disclaimer:** ("disclaimer" is a refusal
keyword), these are compliant answers, not refusals. Only the hard-refusal number is meaningful.Producing and validating this complete format set, BF16 + NVFP4 + a full GGUF ladder, each verified across vLLM, SGLang, and llama.cpp on bleeding-edge Blackwell hardware, was a lot of work. If it's useful to you, I'd genuinely appreciate your support on Patreon 🙏, it keeps releases like this coming.
Apache License 2.0, inherited from the base model by Google. See the official Gemma 4 license page. Uncensoring, format set and validation by Robert Ressl (Hugging Face · Website · LinkedIn · Patreon).