AI Model Release Tracker

Qwen3-Coder-NextOpen Weight

Released

Qwen3-Coder-Next is an AI model released by Qwen on Tuesday, Feb 3 2026, 132 days after Qwen3-Max. It is an open-weight model — the trained weights are available to download and run. It is a 80B parameterThe learned values that shape how a model responds. B means billion and T means trillion. For mixture-of-experts models, the total count is larger than the number active for each token. Parameter count alone does not measure quality. model with a 256k token context windowHow much information the model can work with in one request, measured in tokens — pieces of words or other data. This includes your question, attachments, conversation and reply. Limits can vary by model variant and provider.. Benchmark results (shown below) cover SWE-Bench Pro, SWE-Bench Verified, SWE-Bench Multilingual, and Terminal-Bench 2.0.

Available from

Parasail$0.12$0.80256Kbf16
USD per 1M tokensEvery price here is for one million tokens. A token is roughly three-quarters of a word, so a million tokens is about 750,000 words of text.Checked November 3, 2026Third-party rates from openrouter.ai, not the lab’s own list price.

Benchmarks

Coding

SWE-Bench ProAgentic coding — Can the AI fix real bugs in real software? It's handed actual problems from open-source projects and has to write code that genuinely solves them. Higher is better.
44.3%
#24 of 24
SWE-Bench VerifiedCoding — Real coding tasks pulled from open-source projects — the AI has to find and fix actual bugs. A human-checked version of the original SWE-Bench. Higher is better.
70.6%
#41 of 70
SWE-Bench MultilingualMultilingual coding — Like SWE-Bench, but the coding problems span many programming languages, not just one. Tests how broadly the AI can code. Higher is better.
62.8%
#15 of 19

Terminal & CLI

Terminal-Bench 2.0Agentic terminal coding — Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? (Version 2.0 of the test.) Higher is better.
36.2%
#14 of 17

Compare Qwen3-Coder-Next with

Qwen3-Coder-Next

Suggested comparisons

Qwen3-Coder-NextvsQwen3.8-Max-0902Qwen3-Coder-NextvsGPT-6.1 TerraQwen3-Coder-NextvsClaude Haiku 5.5Qwen3-Coder-NextvsGemini 4 CarbonQwen3-Coder-NextvsMuse Spark OpenQwen3-Coder-NextvsGrok 4.7Qwen3-Coder-NextvsDeepSeek-V4.1-Flash-1017Qwen3-Coder-NextvsMistral Large 4Qwen3-Coder-NextvsKimi K3.1 ProQwen3-Coder-NextvsGLM-5.3-FlashQwen3-Coder-NextvsNemotron 3.5 Bolt E4B 4GB

About

Qwen3-Coder-Next was released in February 2026 as an open-weight model aimed at coding agents and local development. It used an 80-billion-parameter sparse architecture with about 3 billion active parameters per token and a native 262,144-token context. The training focus included multi-step software work, calling tools and recovering when an attempted execution failed.

The released checkpoint operated in non-thinking mode, so its agent capabilities should not be confused with exposing long reasoning blocks in every response. Alibaba documented integration with several command-line and editor-based agent systems. The model provided a self-hosting option for developers building coding workflows, while the surrounding agent harness still supplied tools, executed commands and managed the interaction with a repository.

Frequently asked questions

Qwen3-Coder-Next was released by Qwen on Tuesday, Feb 3 2026.

All Qwen releases

24 tracked