GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. 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
ressl/gemma-4-31B-it-uncensored
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 | ressl/gemma-4-31B-it-uncensored → google/gemma-4-31B-it |
| Quantization | ModelOpt NVFP4_DEFAULT_CFG (dense NVFP4; lm_head + vision tower kept BF16) |
| Size | 20 GB (from 59 GB BF16) |
| Effective refusals | 0/686 across 4 datasets, vLLM and SGLang (base 99/100) |
| Hardware | NVIDIA RTX PRO 6000 Blackwell 96 GB (SM120), driver 610 |
| Toolchain |
The vision tower + embedder are kept in BF16 (vLLM's gemma4_mm loader requires it), so multimodal
input is preserved.
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-NVFP4 \--quantization modelopt --max-model-len 8192 \--enforce-eager --no-enable-flashinfer-autotune --trust-remote-code
--no-enable-flashinfer-autotune avoids a startup hang on SM120.
bash
python -m sglang.launch_server --model-path ressl/gemma-4-31B-it-uncensored-NVFP4 \--quantization modelopt_fp4 --attention-backend triton \--fp4-gemm-backend flashinfer_cutlass --disable-flashinfer-autotune --trust-remote-code
gemma-4 requires --attention-backend triton (it rejects flashinfer). Tested with sglang 0.5.14 on Python 3.12.
***Disclaimer:** prefix, see the BF16 card
for the eval methodology and the full cross-dataset table.preprocessor_config.json, vLLM's multimodal loader fails without it.Producing and validating this complete format set (BF16 + NVFP4 + GGUF, 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 🙏, more at ressl.ch.
Apache License 2.0, inherited from the base model by Google. See the official Gemma 4 license page. Uncensoring, quantization and validation by Robert Ressl (Hugging Face · Website · LinkedIn · Patreon).
| nvidia-modelopt · vLLM 0.23 · SGLang 0.5.14 |