Qwen / text-generation updated 1 year ago

Qwen3-4B-Thinking-2507-FP8

Name: Qwen3-4B-Thinking-2507-FP8
Author: Qwen

Over the past three months, we have continued to scale the thinking capability of Qwen3-4B, improving both the quality and depth of reasoning. We are pleased to introduce Qwen3-4B-Thinking-2507-FP8, featuring the following key enhancements:

Params

4.4 B

Context

262,144

Downloads 30d

205 K

Likes

Commercial use: allowed apache-2.0 Not gated SAFETENSORS View on Hugging Face ↗

Download history

daily snapshots · 10 days

208 K204 K

Jul 28Jul 31Aug 3Aug 6

Can you run it?

Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.

File	Quant	Size	Est. VRAM	Verdict on RTX 4090 · 24 GB
model.safetensors	f8_e4m3	5.2 GB	6.9 GB	✅ Runs comfortably
model.safetensors (bf16, full)	bf16 + 256K ctx	5.2 GB	27.4 GB	❌ Won’t fit

Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.

Run it

copy-paste, exact tags checked against the Hub

~ · curl · api/v1

$ curl -s https://aimodelscomparison.com/api/v1/models/qwen3-4b-thinking-2507-fp8
{
  "hf_id": "Qwen/Qwen3-4B-Thinking-2507-FP8",
  "params_b": 4.41,
  "context_length": 262144,
  "license": { "id": "apache-2.0", "commercial": "yes" },
  "downloads_30d": 204861,
  "vram_estimates": [
    { "quant": "f8_e4m3", "gb": 6.9 }
  ],
  "updated_at": "2026-07-28T18:08:04Z"
}

est. VRAM —on RTX 4090 · 24 GBJSON API →

Specifications

Architecture: Qwen3ForCausalLM
Parameters: 4.4 B
Tensor type: F8_E4M3
Context length: 262,144
Vocabulary: 151,936
Layers / heads: 36 / 32
Licence: apache-2.0
First seen on the Hub: 2025-08-06
Base model: Qwen3-4B-Thinking-2507
Training datasets: undisclosed
Added to our catalog: 2026-07-28

Family

Base model and the most-downloaded derivatives in the catalog.

Base Qwen3-4B-Thinking-2507 Qwen 341 K

Compare with

Qwen3-8B 16.3 M Qwen2.5-1.5B-Instruct 14.0 M Qwen2.5-7B-Instruct 12.1 M Llama-3.1-8B-Instruct 8.0 M Qwen2.5-3B-Instruct 6.0 M Qwen3-4B 4.6 M

Appears in

Best open coding models Commercial-use text models Models that fit in 12 GB