Qwen3 4B FP8 is an open-source language model by Qwen. Features: 4b LLM, VRAM: 5.2GB, Context: 40K, License: apache-2.0, LLM Explorer Score: 0.24.
Arxiv:2309.00071 Arxiv:2505.09388 Autotrain compatible Base model:quantized:qwen/qwen... Base model:qwen/qwen3-4b Conversational Endpoints compatible Fp8 Qwen3 Region:us Safetensors Sharded Tensorflow
Model Card on HF 🤗: https://huggingface.co/Qwen/Qwen3-4B-FP8
Qwen3 4B FP8 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)
Qwen3 4B FP8 Parameters and Internals
| LLM Name | Qwen3 4B FP8 |
| Repository 🤗 | https://huggingface.co/Qwen/Qwen3-4B-FP8 |
| Base Model(s) | |
| Model Size | 4b |
| Required VRAM | 5.2 GB |
| Updated | 2025-10-04 |
| Maintainer | Qwen |
| Model Type | qwen3 |
| Model Files | |
| Model Architecture | Qwen3ForCausalLM |
| License | apache-2.0 |
| Context Length | 40960 |
| Model Max Length | 40960 |
| Transformers Version | 4.51.0 |
| Tokenizer Class | Qwen2Tokenizer |
| Padding Token | <|endoftext|> |
| Vocabulary Size | 151936 |
| Torch Data Type | bfloat16 |
| Errors | replace |
Best Alternatives to Qwen3 4B FP8
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Qwen3 4B Thinking 2507 | 256K / 8.1 GB | 1207168 | 574 |
| GRPO 4 70 | 256K / 8.1 GB | 20 | 0 |
| Qwen3 4B Instruct 2507 | 256K / 8.1 GB | 1473674 | 320 |
| Jan Code 4B | 256K / 8.9 GB | 2270 | 72 |
| Jan V1 4B | 256K / 8.1 GB | 108354 | 353 |
| AgentCPM Explore | 256K / 8.9 GB | 138 | 328 |
| Lightning 4B | 256K / 8.1 GB | 13 | 6 |
| Qwen3 4B Instruct 2507 FP8 | 256K / 5.2 GB | 715358 | 73 |
| DASD 4B Thinking | 256K / 8.1 GB | 443 | 216 |
| Jan V3 4B Base Instruct | 256K / 8.9 GB | 1827 | 60 |
Note: green Score (e.g. "73.2") means that the model is better than Qwen/Qwen3-4B-FP8 .
Rank the Qwen3 4B FP8 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? 52473 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