Ctrl K
Back to models

Alibaba

Qwen3 Next 80B A3B (Reasoning)

  • Released
  • Open weights
  • Apache 2.0
Compare
Intelligence Index

Reasoning setting

11.2
Estimated · Artificial Analysis v4.3.2
API price from
$0.15 in / $1.20 out / 1M tokOpenRouter · Google
Context window
262.14K tok

Performance profile

ReasoningRank #61 of 107
75.9%
GPQA Diamond

Capabilities

Context window

Modalities

Measured setting

Integration

Reasoning trade-offs

Source (opens in a new tab)
Intelligence Index v4.3.2
ReasoningEstimated
Intelligence Index
11.2

Pricing & access

API route
OpenRouter · GoogleAPI
Tier & contextStandard

USD per 1M tokens

Input
$0.15
Output
$1.20

Run locally

GGUF
Context
Estimated memory

~49.9 GiB

RAM + VRAM

Estimate assumptions

One text chat at the selected context, with a 16-bit cache and app allowance. Operating system and image processing are extra; optimized caches can use less.

64 GiB RAMwithout GPU offload

Memory budgetOne text chat
Weights
45.2 GiB
Context + state
0.3 GiB16-bit cache
App allowance
4.5 GiB
Download
48.5 GB
Disk space
~59 GB
Available software
Ollama (opens in a new tab)LM Studio (opens in a new tab)Jan (opens in a new tab)KoboldCpp (opens in a new tab)llama.cpp (opens in a new tab)
Get GGUF (opens in a new tab)
Set up in Ollama · Q4_K_M · 8K context

Save the following as Modelfile. The context matches the estimate above.

FROM hf.co/lmstudio-community/Qwen3-Next-80B-A3B-Thinking-GGUF:Q4_K_M
PARAMETER num_ctx 8192

Then run these commands in your terminal:

ollama create llmcurrent-local -f Modelfile
ollama run llmcurrent-local

Ollama import instructions (opens in a new tab) · For LM Studio or Jan, open the app’s model search, enter lmstudio-community/Qwen3-Next-80B-A3B-Thinking-GGUF, choose Q4_K_M, and set context to 8,192 tokens in the app’s model settings. Software links above open each app’s setup guide.

vLLM setup (opens in a new tab)

Benchmarks

Full leaderboardSource (opens in a new tab)

Humanity’s Last Exam

Frontier reasoning

Qwen3 Next 80B A3B (Reasoning) · Reasoning#82 of 126

  1. 1Claude Opus 5.5Max61.4%
  2. 2Claude Fable 5.1Max59.1%
  3. 3Gemini 4 ArgonHigh57.1%
  5. 79K2 Horizon 3.7BReasoning13.9%
  6. 80Mistral Medium 3.5High13.8%
  7. 81EXAONE 4.5 33BReasoning12.9%
  8. 82Qwen3 Next 80B A3B (Reasoning)Reasoning12.6%
  9. 83Command A+Reasoning12.0%
  10. 84Nova 2.0 LiteHigh11.6%
  11. 85NVIDIA Nemotron 3 Nano 30B A3BReasoning11.4%

Peers use their best measured result on this benchmark.