Qwen3 Coder Next
Alibaba · 1 configuration
Intelligence9.2
Input / 1M$0.35
Output / 1M$1.20
Context256K
Tariff: Alibaba · Source: artificial-analysis · USD per million tokens
Thinking configuration
What does more thinking buy you?
Compare this model’s measured thinking configurations.
Scroll the chart ↔ · tap a point for details
Efficient frontier: best measured score at each cost1 measured configurations · 0 without both values
Selected: thinking default · Click a point to switch configuration
Qwen3-Coder-Next is an open-weight causal language model optimized for coding agents and local development workflows. It uses a sparse MoE design with 80B total parameters and only 3B activated per...
Intelligence IndexArtificial Analysis composite index, not a percentage. Different versions of the index are not directly comparable.
9.2 pointsGPQA DiamondGraduate-level scientific questions. Measures specialist reasoning rather than web-search quality.
73.7 %Humanity’s Last ExamDifficult questions across many disciplines, compared under the same evaluation setup.
10.1 %SciCodeCode generation for scientific research problems.
36.2 %Terminal-Bench 2.1Agent tasks in a terminal environment. Results also depend on the evaluation harness.
38.2 %τ²-BenchTool use and interaction in assistance scenarios. Does not replace an evaluation of your integration.
79.5 %IFBenchVerifiable instruction following. A proxy for precision rather than a benchmark of creative style.
35.2 %Omniscience accuracyFactual accuracy in Artificial Analysis's Omniscience evaluation.
16.2 %Specifications from the catalog
Inputtext
Outputtext
Tool callingYes
Structured outputYes
Speed112.4 tok/s
First token1.23 s
Maximum output235.9K
Catalog listing date2026-02-04