akashml

Qwen: Qwen3.8 27B

Qwen/Qwen3.8-27B
Context

256K

Input

text, image, video

Output

text

Tool calling

Supported

About this model

Qwen3.8-27B is a compact, deployment-friendly dense multimodal model from Alibaba's Qwen3.8 family. Natively supporting text, image, and video inputs, it handles tasks ranging from STEM diagrams and document understanding to long-form video analysis. It features flexible thinking control with adjustable reasoning effort and is designed for autonomous planning, coding, professional work, multimodal reasoning, and long-horizon agentic workflows.

Input modalities

text

Accepted as model input

image

Accepted as model input

video

Accepted as model input

Providers

Available routing options for this model through OneInfer.

Available

Input tokens

$0.450 / 1M tokens

Output tokens

$3.200 / 1M tokens

Routing

OneInfer optimized

Pricing

Current OneInfer pricing for this model.

UsagePrice
Input tokens$0.450 / 1M tokens
Output tokens$3.200 / 1M tokens
Cached input tokens$0.050 / 1M tokens

Performance

Published evaluation results associated with this model.

General agentic

Claw-Eval Avg72.4
Claw-Eval Pass^360.6
QwenClawBench53.4
SkillsBench48.2

Agentic coding

SWE-Bench Verified77.2
SWE-Bench Pro53.5
SWE-Bench Multilingual71.3
TerminalBench 2.059.3
NL2Repo36.2
QwenWebBench1487

Multimodal

MMMU82.9
MMMU-Pro75.8
MathVista mini87.4
DynaMath85.6
VlmsAreBlind97
RealWorldQA84.1
MMStar81.4
MMBench EN-DEV v1.192.3
SimpleVQA56.1

General capabilities and reasoning

MMLU-Pro86.2
MMLU-Redux93.5
SuperGPQA66
C-Eval91.4
GPQA Diamond87.8
Humanity's Last Exam24
LiveCodeBench v683.9
AIME 202694.1
HMMT Feb 202684.3
HMMT Nov 202590.7
IMOAnswerBench80.8

Document understanding

CharXiv RQ78.4
CC-OCR81.2
OCRBench89.4

Spatial intelligence

ERQA62.5
CountBench97.8
RefCOCO Avg92.5
EmbSpatialBench84.6
RefSpatialBench70

Video understanding

VideoMME87.7
VideoMMMU84.4
MLVU86.6
MVBench75.5

Visual agent

V*94.7
AndroidWorld70.3

API example

curl https://api.oneinfer.ai/v1/ula/chat/completions \
  -H "Authorization: Bearer $JWT_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "Qwen/Qwen3.8-27B",
    "messages": [
      { "role": "user", "content": "Hello!" }
    ]
  }'