Qwen3 4B 2507 Instruct
Analysis Summary
Qwen3 4B 2507 Instruct is an Alibaba instruction-tuned model released in August 2025. The supplied results include a 6.7 intelligence index, math index of 52.3, MMLU-Pro result of 0.672, and instruction-following result of 0.335. These task measurements indicate useful performance on some structured tests, but the low overall index and limited terminal benchmark result argue against relying on it for demanding coding or tool-driven work.
For business use, it could be tested on constrained extraction, straightforward question answering, or basic structured responses, with human review for consequential outputs. The data lists no price, modality, or context window, so cost-effectiveness and document-handling capacity cannot be assessed. It is not a preferred model for client-facing writing that requires strong tone control or complex autonomous workflows.
Assessed September 23, 2026
Editorial notes
Qwen3 4B 2507 Instruct has measured math and instruction-following results, but its low overall capability and limited tool-related results make it better suited to bounded tasks.
The DFO score ranks models on the Artificial Analysis Intelligence Index and is refreshed as new models launch. The write-up and verdict are our own. Feedback?
DFO Verdict
Qwen3 4B 2507 Instruct has measured math and instruction-following results, but its low overall capability and limited tool-related results make it better suited to bounded tasks.
Benchmarks
Artificial Analysis data refreshed Oct 23, 2026. Arena ratings from the Oct 20, 2026 text leaderboard, with style control. Bars: magenta = reasoning, ink = coding and agents, cyan = instructions and long context.
Artificial Analysis has not published coding or agentic results for Qwen3 4B 2507 Instruct yet. Until it does, the model sits below tested models on the Coding and AI Agents lists.
How Qwen3 4B 2507 Instruct compares
Qwen3 4B 2507 Instruct ranks #330 of 443 AI models we track for overall intelligence. Qwen3 4B 2507 Instruct is currently free to use via OpenRouter.
Dark bar = input · light bar = output, scaled to the priciest peer.
1M tokens ≈ 8 full-length novels or ~2,500 pages of business documents in a single request.
Frequently asked questions about Qwen3 4B 2507 Instruct
How much does Qwen3 4B 2507 Instruct cost?
Qwen3 4B 2507 Instruct is currently available for free via OpenRouter.
Who created Qwen3 4B 2507 Instruct?
Qwen3 4B 2507 Instruct is developed by Alibaba and was released on August 6, 2025.