Qwen

Qwen3.8 Max (0902)

by Qwentext+image+video->text1 endpoint
Neura Intelligence Index
87.8/ 100
Rank
#19
Confidence
medium
Benchmarks
7(18% coverage)

Overview

Qwen3.8 Max 0902 is an updated snapshot of Qwen3.8 Max from Alibaba's Qwen team. It is a 2.4-trillion-parameter mixture-of-experts model that accepts text, image, and video input and returns text, with a 1M-token context window and reasoning enabled by default.

This snapshot is post-trained for coding and agentic work, including multi-step software projects, multi-tool orchestration, and long-horizon task execution. It also targets chart reasoning, document parsing, and multimodal understanding over long documents and extended video. Tool calling, structured outputs, and configurable reasoning effort are supported.

Capabilities

Text generation
Image understanding
Video understanding
Tool use / Function calling
Structured output
Extended reasoning
Prompt caching
Long context
Top-P sampling
Stop sequences
Deterministic seed
Log probabilities

Modalities

Input
TextImageVideo
Output
Text

Technical Specifications

Context Window
1M tokens
Max Output
131.1K tokens
Tokenizer
Qwen
Uptime (24h)
10000.0%
Parameters
2400000000000B
License
qwen3_8_maxOpen Weights
Country
🇨🇳CN
Release Date
August 2, 20261mo

Supported Parameters

frequency_penaltyinclude_reasoninglogprobsmax_tokenspresence_penaltyreasoningreasoning_effortresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p

Benchmark Performance

general

BenchmarkScoreSource
MRCR v2
92.9
verified

coding

BenchmarkScoreSource
SWE-Bench Pro
67.7
verified

reasoning

BenchmarkScoreSource
GPQA Diamond
92.6
verified
Humanity's Last Exam
43.6
verified

multimodal

BenchmarkScoreSource
ScreenSpot Pro
84.5
verified
MMMU-Pro
82.3
verified

agent

BenchmarkScoreSource
Toolathlon
72.5
verified

Performance Over Time

Pricing

Per 1M tokens
Input$2.00
Output$6.00
Cache read$0.25
Cache write$2.50
Added
September 3, 2026
Last synced 9/20/2026