GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
axteam
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
axteam
Model Tree
Qwen/Qwen3.5-4B
Input Modalities
Output Modalities
Supported Functionality
nvfp4-pack-quantized.model.visual.*), the multi-token-prediction head (mtp.*), all RMSNorms, token embeddings, the LM head, the entire linear-attention (GatedDeltaNet/SSM) path — FP4 is too coarse for the SSM weights, so the full linear_attn.* stays BF16. Left unquantized to preserve quality (about 44% of all parameters in this build).chat_template / generation_config are identical to the original. config.json is the original plus a quantization_config block.This is post-training quantization (PTQ) — no retraining. AWQ uses a small calibration set (no gradient) to protect salient weight channels before rounding.
Every column was measured with the same harness and settings — lm-evaluation-harness on vLLM, greedy, few-shot completion (no chat template). Percentages are recovery vs. the BF16 original.
| Benchmark | BF16 (original) | FP8-dynamic | AWQ-8bit | NVFP4A16 (this repo) | NVFP4 | AWQ-4bit |
|---|---|---|---|---|---|---|
| MMLU (5-shot) | 55.18 | 57.99 (105.1%) | 54.79 (99.3%) | 60.58 (109.8%) | 58.61 (106.2%) | 60.89 (110.3%) |
| GSM8K (5-shot, strict) | 74.07 | 73.92 (99.8%) | 73.84 (99.7%) | 76.72 (103.6%) | 73.31 (99.0%) | 76.42 (103.2%) |
| ARC-Challenge (0-shot) | 54.61 | 53.92 (98.8%) | 54.52 (99.8%) | 54.86 (100.5%) | 53.92 (98.8%) | 54.35 (99.5%) |
| Winogrande (5-shot) | 68.75 | 67.96 (98.9%) | 69.14 (100.6%) | 68.90 (100.2%) | 69.38 (100.9%) | 68.90 (100.2%) |
| KMMLU (5-shot, direct) | 50.47 | 50.13 (99.3%) | 50.48 (100.0%) | 51.49 (102.0%) | 50.26 (99.6%) | 49.02 (97.1%) |
| HAE-RAE (0-shot) | 62.42 | 61.50 (98.5%) | 62.05 (99.4%) | 61.69 (98.8%) | 59.76 (95.7%) | 58.48 (93.7%) |
| Average | 60.92 | 60.90 (100.0%) | 60.81 (99.8%) | 62.37 (102.4%) | 60.87 (99.9%) | 61.34 (100.7%) |
Fidelity to the BF16 original on an internal mixed English/Korean set (100 prompts, BF16 greedy answers scored under each build — same harness for every row).
| Metric | This model | BF16 original |
|---|---|---|
| Perplexity (lower = closer) | 1.6374 | 1.5703 (+4.3%) |
| Top-1 next-token agreement | 91.7% | 99.2% |
These measure fidelity to the base model, not absolute task scores.
| Runtime | vLLM — the architecture is in vLLM ≥ 0.17.0; tested here with vLLM 0.32.0. |
| vLLM install | vLLM 0.32 itself needs Python 3.11–3.14, torch 2.13 (CUDA 13.0 build, so an NVIDIA driver with CUDA 13 support), transformers ≥5.16.1,<5.20 and compressed-tensors 0.17.0 (both pinned by vLLM; 0.17.0 reads this checkpoint, which was saved with 0.19.0 — do not upgrade it separately). |
| GPU | NVIDIA Turing (sm75) or newer — weight-only FP4, no FP4 tensor cores needed. V100 (sm70) is not supported. (Kernel choice per GPU from vLLM 0.32's selection code; only B200 was run here.) |
| Memory | Weights: 5.6 GB. Add KV cache (grows with context length × concurrent requests) and a few GB for activations / CUDA graphs; use --tensor-parallel-size N when the weights do not fit one GPU. |
| Transformers (without vLLM) | transformers>=5.2.0 + compressed-tensors>=0.15.0 (checked on CPU with Qwen3.5-0.8B FP8 / AWQ-4bit / NVFP4A16 / NVFP4). |
| trust_remote_code | not needed |
Start an OpenAI-compatible server:
bash
vllm serve axteam/Qwen3.5-4B-NVFP4A16 --max-model-len 32768Query it:
bash
curl http://localhost:8000/v1/chat/completions -H "Content-Type: application/json" -d '{ "model": "axteam/Qwen3.5-4B-NVFP4A16", "messages": [{"role": "user", "content": "대한민국의 수도는 어디인가요?"}], "max_tokens": 512}'Offline inference in Python:
python
from vllm import LLM, SamplingParams llm = LLM(model="axteam/Qwen3.5-4B-NVFP4A16", max_model_len=32768)params = SamplingParams(max_tokens=1024)out = llm.chat([{"role": "user", "content": "Explain model quantization in two sentences."}], params)print(out[0].outputs[0].text) # image input (URL or data: URI)messages = [{"role": "user", "content": [ {"type": "image_url", "image_url": {"url": "https://example.com/your-image.png"}}, {"type": "text", "text": "Describe this image."},]}]out = llm.chat(messages, params)print(out[0].outputs[0].text)Thinking mode is on by default in the chat template; to turn it off pass
chat_template_kwargs={"enable_thinking": False}(Pythonllm.chat(..., chat_template_kwargs=...), or"chat_template_kwargs": {"enable_thinking": false}in the request body). Text-only serving: add--limit-mm-per-prompt '{"image": 0, "video": 0}'to skip loading the vision encoder's memory budget.
The KV cache is kept in BF16 by default. KV-cache precision is a runtime choice, independent of the weight quantization above; for long contexts or large batches FP8 roughly halves KV-cache memory:
bash
vllm serve axteam/Qwen3.5-4B-NVFP4A16 --kv-cache-dtype fp8Quantized derivative of Qwen/Qwen3.5-4B (Apache-2.0). All original weights, tokenizer and chat template belong to the Qwen team.
Qwen/Qwen3.5-4B 를 NVFP4A16 (W4A16, FP4 weights) 로 양자화한 서빙용(4-bit) 모델입니다. 작성: axteam.
nvfp4-pack-quantized.vllm serve axteam/Qwen3.5-4B-NVFP4A16 --max-model-len 32768 로 OpenAI 호환 서버(포트 8000), 또는 Python LLM(model="axteam/Qwen3.5-4B-NVFP4A16").chat(...) (위 예시). thinking 끄기: chat_template_kwargs={"enable_thinking": False}. 이미지 입력은 image_url 메시지로.--kv-cache-dtype fp8.curl http://localhost:8000/v1/chat/completions -H "Content-Type: application/json" -d '{ "model": "axteam/Qwen3.5-4B-NVFP4A16", "messages": [{"role": "user", "content": "대한민국의 수도는 어디인가요?"}], "max_tokens": 512}'from vllm import LLM, SamplingParams llm = LLM(model="axteam/Qwen3.5-4B-NVFP4A16", max_model_len=32768)params = SamplingParams(max_tokens=1024)out = llm.chat([{"role": "user", "content": "Explain model quantization in two sentences."}], params)print(out[0].outputs[0].text) # image input (URL or data: URI)messages = [{"role": "user", "content": [ {"type": "image_url", "image_url": {"url": "https://example.com/your-image.png"}}, {"type": "text", "text": "Describe this image."},]}]out = llm.chat(messages, params)print(out[0].outputs[0].text)