Qwen3-Coder-Next

Qwen3-Coder-Next ranks first on 3 of the 31 benchmarks with five or more models scored.

Qwen Released February 2nd, 2026 Family Qwen3-Coder The lab's model card Compare with others

Benchmarks with a score

Ranked first, out of five or more models

In the top three

Qwen Feb 2026

Qwen3-Coder-Next

Open weights Small

138.5 #190 of 404

Self-Reported Index, from the lab's own numbers. No Epoch Capabilities Index score yet.

SuperGPQA 57.5% #22 /35
FullStackBench 60.6% #5 /8
Bird-SQL 63.6% #1 /13

Where it ranks

Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.

Knowledge and language

Benchmark	Score	Rank	Source
MMLU	87.7%	#30 of 131	Lab model card
MMLU-Pro	80.5%	#77 of 167	Lab model card
MMLU-Redux	91.2%	#22 of 45	Lab model card
C-Eval	88.2%	#25 of 46	Lab model card
SuperGPQA	57.5%	#22 of 35	Lab model card

Math and science

Benchmark	Score	Rank	Source
AIME 2024	89%	#22 of 88	Lab model card
GPQA Diamond	74.5%	#151 of 302	Lab model card
AIME 2025	83.1%	#81 of 143	Lab model card
HMMT 2025	70.2%	#47 of 54	Lab model card
HMMT Nov 2025	75.6%	#26 of 28	Lab model card

Coding

Benchmark	Score	Rank	Source
MultiPL-E	88.2%	#1 of 18	Lab model card
Bird-SQL	63.6%	#1 of 13	Lab model card
CRUXEval	95.9%	#1 of 9	Lab model card
Spider	83.7%	#1 of 4	Lab model card
EvalPlus	86.6%	#2 of 6	Lab model card
Aider Polyglot	66.2%	#12 of 30	Lab model card
Codeforces rating	2,100	#19 of 47	Lab model card
SWE-bench Verified	70.6%	#74 of 155	Lab model card
FullStackBench	60.6%	#5 of 8	Lab model card
OpenHands Index	43.8%	#24 of 33	Official leaderboard
SWE-bench Multilingual	62.8%	#57 of 73	Lab model card
LiveCodeBench v6	58.9%	#76 of 90	Lab model card
WeirdML	34.4%	#95 of 112	Official leaderboard
SWE-Bench Pro	44.3%	#77 of 81	Lab model card
OJBench	23%	#16 of 16	Lab model card
ts-bench v1	92%	#9 of 9	Official leaderboard

Agents and tools

Benchmark	Score	Rank	Source
SnakeBench	21	#124 of 188	Official leaderboard
Terminal-Bench 2.0	36.2%	#55 of 63	Lab model card
Era by Eon	55.6%	#8 of 9	Official leaderboard
ClawProBench	46.8%	#46 of 49	Official leaderboard
Template Following generalization (Qwen)	92.7%	#1 of 1	Lab model card

Safety and security

Benchmark	Score	Rank	Source
nocot-bench (NCRI)	65.5	#118 of 185	Official leaderboard
SusVibes	7.50%	#10 of 10	Official leaderboard
PrimeVul	48.3%	#2 of 2	Lab model card
CWEval	56.3%	#1 of 1	Lab model card

Its lab's numbers put it at 138.5, number 190 of 404

The Self-Reported Index fits one score to the 21 capability results Qwen reported for this model, on the scale of the Epoch Capabilities Index. Labs choose which benchmarks to show and how to run them, so read it as the lab's claim, not an independent measurement. The 90% interval runs from 136.6 to 140.4. How we compute it.