skip to content
gigarouter gigarouter

Qwen3 4B Instruct 2507

Qwen/Qwen3-4B-Instruct-2507

published Aug 2025 · updated Sep 2025

Qwen3 4B Instruct 2507 is a text-generation model that is an updated version of Qwen3-4B non-thinking mode, with significant improvements in instruction following, logical reasoning, coding, tool usage, and multilingual support.

price
$0.05
/ 1M tokens

specs

Tasktext-generation
ArchitectureCausal Language Model with GQA (32 Q heads, 8 KV heads)
Parameters4.0B total (3.6B non-embedding)
Context Length262,144 tokens natively
LicenseApache 2.0

about this model

Qwen3-4B-Instruct-2507 is a causal language model for text generation that offers significant gains over its predecessor in instruction following, logical reasoning, mathematics, science, coding, tool usage, and multilingual capabilities. It supports a native context length of 262,144 tokens, extendable to 1,000,000 tokens. This model operates exclusively in non-thinking mode, meaning it does not generate thinking response blocks in its output. It is released under the Apache 2.0 license.

Key Capabilities and Improvements

Substantial improvements have been made in long-tail knowledge coverage across 119 languages and dialects. Alignment with user preferences in subjective and open-ended tasks is markedly better, enabling more helpful responses and higher-quality text generation. The model also demonstrates enhanced 256K long-context understanding.

Benchmark Performance

The model achieves strong results across knowledge, reasoning, coding, alignment, and agentic tasks. Selected scores include:

  • Knowledge: MMLU-Pro 69.6, GPQA 62.0
  • Reasoning: AIME25 47.4, LiveBench (20241125) 63.0
  • Coding: LiveCodeBench v6 35.1, MultiPL-E 76.8
  • Alignment: Arena-Hard v2 43.4, Creative Writing v3 83.5
  • Agent: BFCL-v3 61.9, TAU1-Retail 48.7
  • Multilingual: MultiIF 69.0, PolyMATH 31.1
Radar chart comparing Qwen3-4B-Instruct-2507 performance against GPT-4.1-nano, Qwen3-30B-A3B, and Qwen3-4B across benchmark categories.

This model is hosted by gigarouter as an OpenAI-compatible API, providing convenient access for text-generation workloads without the need for manual deployment.

best for

FAQ

What is this model best for?

It excels at instruction following, logical reasoning, coding, tool usage, and multilingual tasks, with improved alignment and higher-quality text generation.

How does it compare to the original Qwen3-4B non-thinking model?

This updated version shows significant gains across benchmarks, including MMLU-Pro (58.0 to 69.6), AIME25 (19.1 to 47.4), and Arena-Hard v2 (9.5 to 43.4).

What is the license for this model?

It is released under the Apache 2.0 license.

What is the input and output format?

It uses the standard chat template with user/assistant messages. The model supports non-thinking mode only, so no thinking blocks are generated.

How can I call this model via the API?

Use the gigarouter OpenAI-compatible endpoint with your API key. The endpoint supports chat completions; pass the model name as qwen3-4b-instruct-2507.

call it
# rerank documents by relevance; billed per document
curl https://gigarouter.ai/v1/rerank \
  -H "Authorization: Bearer $GR_KEY" \
  -d '{"model":"Qwen/Qwen3-4B-Instruct-2507","query":"capital of France",
       "documents":["Paris is the capital of France.","Bananas are yellow."]}'

try it live

runs the real hosted model on a shared demo allowance · get your own key + $25 free →

related text generation models

compare all →