Qwen3.8 27B Thinking

Qwen3.8 27B is an open-weight multimodal model from Alibaba for reasoning, coding, visual understanding, tool use, and structured output. This variant enables thinking by default. Requests containing video cost $0.32 per million input tokens and $2.50 per million output tokens.

Reasoning
Vision
Video Input
Tool Calling
Structured Output

Added Aug 16, 2026

Model weights

Pricing

Auto routing · per 1M tokens

Input: $0.15
Output: $0.70
Cache read: $0.040

Compare provider prices

Specifications

Context window: 1M
Max output: 32.8K
Parameters: 27B
Avg output (7d): 1.6K tokens; Longer than 87% of models

Benchmarks

Sourced from Artificial Analysis.

Intelligence Index

33.7

Better than 87% of models compared

Coding Index

68.1

Better than 78% of models compared

Agentic Index

45.8

Better than 86% of models compared

Agentic work

AutomationBench-AA

Workflow automation with guardrail penalties

48.2%

Better than 64% of models compared

AutomationBench-AA Tasks Completed

Fully completed workflows without guardrail violations

15.7%

Better than 17% of models compared

AA-Briefcase

Agentic knowledge work (Elo)

1397 Elo

Better than 77% of models compared

GDPval-AA v2

Economically valuable tasks (Elo)

1423 Elo

Better than 77% of models compared

Document reasoning

GDP.pdf

Professional PDF reasoning: all-pass rate

16.6%

Better than 60% of models compared

AA-LCR v1.1

Long context reasoning with updated grading

82.0%

Better than 92% of models compared

MLCR-AA

Medical long-context reasoning

21.7%

Better than 73% of models compared

Reasoning

HLE

Humanity's Last Exam

33.9%

Better than 81% of models compared

CritPt

Research-level physics reasoning

5.4%

Coding

Terminal-Bench v4.0

Practical coding and terminal tasks

5.6%

Better than 57% of models compared

SciCode

Python programming for scientific computing

46.6%

Better than 41% of models compared

Knowledge

AA-Omniscience Accuracy

Proportion of correctly answered questions

15.6%

AA-Omniscience Hallucination Rate

Rate of incorrect answers among non-correct responses

30.3%

Legacy benchmarks

GPQA Diamond (legacy)

Graduate-level scientific reasoning

90.5%

Better than 90% of models compared

AA-LCR (unversioned / legacy)

Long context reasoning evaluation

82.0%

Better than 92% of models compared

GDPval-AA (unversioned / legacy)

Economically valuable tasks

46.0%

Last updated Oct 14, 2026

Artificial Analysis

Providers

Choose explicit providers for this model. Auto routing remains available as the default option.

Loading provider options…