Qwen3 Next 80B A3B Thinking

Sep 2025
Qwen3 Next 80B A3B Thinking
Compare in
on
—per task?
$0.50 / $6per 1M in / out
153 tok/sfirst token in 0.7 s?
262Kcontext

Rankingin thinking mode?

Strong at

B
Reasoning#160 of 336Science#146 of 312Long context#149 of 314Instructions#84 of 245
C
Agents & work tasks#146 of 177Tool calling#125 of 242Knowledge#217 of 320

Alternatives?

ModelScoreCostPer taskWhat changes
CheaperGemini 3 Flash Previewthinking+15%-40%$0.50 / $3BetterGPT OSS 20BhighOpen+6%-98%$0.011Context: 262K → 131KOpenGPT OSS 120Bhigh+11%-97%$0.11Context: 262K → 131K+1CheaperGLM 4.7thinkingOpen+13%-47%$0.60 / $2.20Context: 262K → 203K+1CheaperKimi K2 ThinkingthinkingOpen+12%-55%$0.47 / $2Context: 262K → 256K+1

Run it yourself

estimate
Fits on✗ 8 GB (phone, laptop)✗ 16 GB✗ 24 GB GPU✗ 48 GB GPU✓ 80 GB GPU · compressed
LicenceApache-2.0free for commercial use

Benchmarks

all categories
Chat & Knowledge
Humanity's Last Exam12.6#161
AA Omniscience Index-51.4#218
CritPt (physics)0#182
AA Intelligence Index 4.3.2≈11.2#196
AA Omniscience · Accuracy19#167
AA Omniscience · Hallucination Rate87#225
AA-LCR (long context) 1.163.7#149
GPQA Diamond75.9#147
AIME 2025 · AA archived84.3#33
Coding
Terminal-Bench Hard (version unstated)9.8#127
Terminal-Bench 2.16.7#131
LiveCodeBench · AA archived78.4#20
Agents & Tool Use
GDPval-AA (work tasks) 2.10#146
τ³-Banking · AA6.2#108
τ²-bench · AA41.5#127
Writing & Language
IFBench (instructions)60.7#84

FAQ

How much does Qwen3 Next 80B A3B Thinking cost?›

$0.50 per 1M input tokens and $6 per 1M output tokens.

Is Qwen3 Next 80B A3B Thinking free to use?›

There is no free API; the cheapest paid price is $0.50 per 1M input tokens. The weights are open, so you can run it yourself.

How good is Qwen3 Next 80B A3B Thinking, and what is it best at?›

#160 of 336 in Chat & Knowledge, #126 of 239 in Coding, #146 of 177 in Agents & Tool Use, #120 of 197 in Writing & Language.

Is Qwen3 Next 80B A3B Thinking better than Seed OSS 36B Instruct?›

On AIME 2025 · AA archived, Qwen3 Next 80B A3B Thinking scores 84.3 vs 84.7 — about the same.

How can I use Qwen3 Next 80B A3B Thinking?›

Through the Alibaba (Qwen) API.

What is the context window of Qwen3 Next 80B A3B Thinking?›

262K tokens, with up to 33K output tokens.

When was Qwen3 Next 80B A3B Thinking released? Is there a newer version?›

Released 2025-09-09. It is the current version in its line.

Is Qwen3 Next 80B A3B Thinking open source?›

Yes, the weights are open (Apache-2.0 licence).

Can I run Qwen3 Next 80B A3B Thinking locally?›

The weights are open and can be self-hosted; hardware needs are not confirmed yet.

Can Qwen3 Next 80B A3B Thinking read images?›

No, it accepts text only.

What is Qwen3 Next 80B A3B Thinking?›

Qwen3 Next 80B A3B Thinking is an open-weights language model from Alibaba (Qwen), released Sep 2025.

What are the best alternatives to Qwen3 Next 80B A3B Thinking?›

Gemini 3 Flash Preview (cheaper), GPT OSS 20B (better), GPT OSS 120B (open weights).