Private AI

Nano GPT logo
NanoGPT
Nano GPT logo
NanoGPT
Nano GPT logo
NanoGPT
Search
 
 
 
 
All conversations

Create

Chat
Studio

Explore

Models
Pricing
AI Apps
Blog

Library

Gallery
Workspace
Usage & billing
API
Suggestions and bugs
Use cases

|

Terms

|

Privacy

|

Refunds
Usage & billing
Teams
Referrals
Presets
Settings
FAQPress & mediaBrand assetsBenchmarksNano RPCEarn crypto
  1. Models
  2. Text
  3. Qwen

Qwen3.8 Max

Qwen3.8 Max is Qwen's 2.4T-parameter flagship model for coding, knowledge work, full-stack development, data analysis, and long-running agent workflows in non-thinking mode. It supports text, image and PDF input, tool calling, structured output, and a near-million-token context window.

  • Vision
  • Native PDF input
  • Tool Calling
  • Structured Output

Added Aug 3, 2026

Pricing

Auto routing · per 1M tokens
Input
$2.00
Output
$6.00
Cache read
$0.25
Compare provider prices

Specifications

Context window
991K
Max output
65.5K
Avg output (7d)
668 tokens
Longer than 52% of models

Benchmarks

Sourced from LMArena.

Arena Score

1484.0

Overall Rank

#23 / 421

Votes

36,324

Confidence Interval

1479.4 - 1488.6

Category Scores

Coding

#31 / 416

9,706 votes

1523.9

Math

#10 / 404

1,804 votes

1504.5

Longer Query

#22 / 399

15,949 votes

1495.2

Creative Writing

#16 / 419

7,414 votes

1471.5

Instruction Following

#27 / 421

12,743 votes

1477.5

Hard Prompts

#24 / 421

23,736 votes

1506.1

Additional Categories
22

Industry Medicine And Healthcare

#4 / 390

2,357 votes

1519.7

Chinese

#12 / 395

2,715 votes

1540.9

Industry Life And Physical And Social Science

#12 / 419

5,511 votes

1515.3

English

#15 / 421

13,500 votes

1493.4

Spanish

#16 / 303

1,206 votes

1480.1

Multi Turn

#17 / 419

5,502 votes

1494.7

French

#18 / 301

1,304 votes

1505.6

Industry Entertainment And Sports And Media

#18 / 419

9,140 votes

1463.7

Hard Prompts English

#20 / 419

8,932 votes

1510.7

Industry Writing And Literature And Language

#21 / 420

9,747 votes

1475.0

Korean

#21 / 289

639 votes

1449.0

Exclude Ties

#23 / 421

26,953 votes

1497.8

Industry Business And Management And Financial Operations

#24 / 414

6,551 votes

1484.9

Non English

#24 / 421

22,824 votes

1472.9

Expert

#25 / 371

3,788 votes

1514.7

Russian

#26 / 385

5,085 votes

1490.5

Industry Legal And Government

#28 / 393

2,660 votes

1494.7

Industry Mathematical

#28 / 401

2,064 votes

1496.0

Industry Software And It Services

#30 / 421

14,030 votes

1513.8

Japanese

#32 / 288

622 votes

1464.6

Polish

#37 / 244

559 votes

1479.2

German

#40 / 320

561 votes

1475.8

Published 2026-11-09 · Matched as Qwen3.8-Max · CC BY 4.0

LMArena Dataset

Leaderboard results

BenchmarkScoreRank on NanoGPT
Epoch AI Benchmarking Hub

Epoch AI data, CC BY 4.0 · Updated Nov 12, 2026

Epoch Capabilities Index

157.6

#28 of 115

FrontierMath v2 (Tiers 1–3)

74.7%

#24 of 83

GPQA Diamond

92.7%

#25 of 119

SimpleQA Verified

45.8%

#39 of 82

FrontierMath v2 (Tier 4)

46.3%

#26 of 67

OTIS Mock AIME 2024–2025

99.4%

#18 of 121

MirrorCode

52.2%

#17 of 39

Chess Puzzles

46.0%

#27 of 82

Mystery Game Puzzles

38.0%

#21 of 77

EQ-Bench

EQ-Bench data, MIT · Updated Nov 12, 2026

Creative Writing v3

1844

#19 of 96

Filtered and ranked to models on NanoGPT.

Providers

Choose or restrict providers with the API. Provider-selection prices include the selection markup. Provider selection guide · Pricing API (JSON, USD per 1K tokens)

Cache write /M
Auto
Auto (ZDR)
$2.00$6.00$0.25$2.50N/AN/A
Auto
Auto (Anonymized)
$2.00$6.00$0.25$2.50N/AN/A
SCX (EU only)
SCX (EU only)
EU
$1.62$4.86$0.208N/AN/AN/A
SCX (Global)
SCX (Global)
GLB
$1.62$4.86$0.208N/AN/AN/A
Alibaba
Alibaba
SG
$2.10$6.30$0.263$2.63N/AN/A
Fireworks
Fireworks
US
$2.10$6.30$0.263N/AN/AN/A
Novita
Novita
US
$2.10$6.30$0.263N/AN/AN/A
SiliconFlow
SiliconFlow
FP8
SG
$2.10$6.30$0.263N/AN/AN/A
Venice
Venice
20% off
US
$2.63$2.10$7.88$6.30$0.328$0.263N/AN/AN/A
Uomi
Uomi
BF16
GLB
$2.21$6.62N/AN/AN/AN/A
TensorX
TensorX
EU
$2.63$6.30$0.656N/AN/AN/A
Privacy:ZDR, prompts not retainedNo training on promptsRetention and training unclear

Related text models

Compare Qwen3.8 Max with similar models from the same provider or model family.

Qwen3.8 Max Prime

qwen/qwen3.8-max-prime

Qwen3.8 Max Prime is a higher-throughput variant of Qwen3.8 Max for coding, complex analysis, and long-running agent workflows. It accepts text, image, and video input, supports tool calling and structured output, and has a 1M-token context window. Reasoning is always enabled.

Qwen3.8 Max 0902

qwen/qwen3.8-max-0902

Qwen3.8 Max 0902 is Alibaba's September 2 checkpoint of its flagship Qwen3.8 Max model for coding, knowledge work, data analysis, and long-running agent workflows. It supports text, image, video, PDF input, selectable thinking, tool calling, structured output, and a near-million-token context window.

Qwen3.8 Max Thinking

qwen/qwen3.8-max:thinking

Qwen3.8 Max Thinking enables generation-time reasoning for deeper coding, knowledge work, data analysis, and long-running agent workflows. It supports text, image, video, PDF input, tool calling, structured output, and a near-million-token context window.

Qwen3.7 Max

qwen/qwen3.7-max

Qwen3.7 Max is built for agentic coding, office automation, and long-running tool workflows in non-thinking mode.

Qwen3.7 Max Thinking

qwen/qwen3.7-max:thinking

Qwen3.7 Max with thinking mode enabled for deeper reasoning, coding, and long-running tool workflows.

Qwen3.6 Max Preview

qwen/qwen3.6-max-preview

Qwen3.6 Max Preview is Alibaba's flagship Qwen 3.6 model for complex tasks. It supports both thinking and non-thinking modes in a single model id.