Skip to content
llmwaves

Qwen3 Coder Next

Qwen · released Feb 4, 2026

Open weightsTool useStructured output

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...

Specification

Context window
262K
Max output
236K
Knowledge cutoff
Not stated
Parameters
79.7B
Licence
apache-2.0
Serving providers
1
Moderated
No
Uptime
100.0%

Intelligence

9.2

38th percentile

Coding

36.2

Coding Index

Agentic

0.9

Agentic Index

Output speed

60 t/s

Median across providers

Latency

540ms

Time to first token

Cost per task

$0.03

Estimated

Benchmarks

Where the score comes from

The Intelligence Index is a composite. These are the underlying evaluations this model was actually measured on.

Evaluation scores

Percentage correct · higher is better

  • τ²-bench (Telecom)
    79.5%
  • GPQA Diamond
    73.7%
  • AA-LCR (long context)
    47.0%
  • SciCode
    36.2%
  • IFBench
    35.2%
  • Terminal-Bench Hard
    18.2%
  • Humanity's Last Exam
    10.1%

An evaluation missing from this list was not run for this model — it is not a zero.

View as table
EvaluationScore
τ²-bench (Telecom)79.5%
GPQA Diamond73.7%
AA-LCR (long context)47.0%
SciCode36.2%
IFBench35.2%
Terminal-Bench Hard18.2%
Humanity's Last Exam10.1%

Against its peers

Intelligence Index · this model highlighted, nearest peers in grey

Peers are the models sitting closest on the Intelligence Index — the set you would realistically choose between.

View as table
ModelIntelligence
R1 Distill Llama 70B9.8
Qwen3 Next 80B A3B Instruct9.6
Qwen3 Coder 30B A3B Instruct9.6
Llama 4 Maverick9.3
Qwen3 Coder Next9.2
Mistral Medium 3.19.2
gpt-oss-20b9.0
Mistral Medium 39.0

Percentile among all indexed models

Intelligence38th
Coding41th
Agentic0th
Terminal work57th

Pricing

What it costs to run

List prices per million tokens, plus what one representative task works out to.

List price

Input / 1M tokens
$0.12
Output / 1M tokens
$0.8
Cached input / 1M
$0.07
Blended 3:1
$0.29

One task, estimated

$0.03

Input tokens
50,000
Output tokens
25,000
Profile
Standard

Estimated from list pricing: 50K input tokens plus 80K output tokens for reasoning models (25K for non-reasoning).

Serving providers

Speed and latency figures are medians across these providers, so a widely-served model reports a blend rather than any single endpoint.