Skip to content

Alibaba · Chat

Qwen: Qwen3 VL 8B Thinking

Alibaba 🇨🇳ChatQwen family
llminfo IQ
109.0
Market median price
$0.25 blended / 1M
Max context
256K
Listings
4

Evidence and sources

International storefront (alibabacloud.com) is Singapore-incorporated; mainland operating entity 杭州阿里云智能科技有限公司 (Hangzhou).

🤗 Qwen/Qwen3-VL-8B-Thinkingllminfo IQ 109.0, ranked among chat models. From 103 benchmarks. Plenty of data behind it.How llminfo IQ is computedPoor value (41.1). Costs more per task than its benchmark results predict. From 1 benchmark with measured task costs. Little data behind it, so treat it as a rough estimate.How Value for money worksReleased 11 Oct 2025Maker HQ: CN (source)
Sources: GenAI Prices data, FlowDot data, Hugging FacePrimary data from GenAI Prices from FlowDot: name from Hugging Face: release date

Advantages and trade-offs

Compared with 10 chat models, picked mainly by popularity and IQ
We only make a claim when there's enough data on models of the same type. Updated 22 Nov, 06:07 UTC.Prices come from the model's maker where they sell it, otherwise their official partner, otherwise the market median. This comparison uses 7 maker's prices; 3 market median prices.

Input & output

as stated by the model's providers
Text
Images
Text

Capabilities

Supported operations and provider-stated features

VisionFunction calling

Listings

 302aiQwen/Qwen3-VL-8B-Thinking · thinkingSource: 302.ai data256K$0.072$0.715
 Flowdotqwen/qwen3-vl-8b-thinking · thinkingSource: FlowDot data131K$0.117$1.365
 OpenRouterqwen/qwen3-vl-8b-thinking · thinkingSource: GenAI Prices dataNot published by the provider$0.117$1.365
 Llamagatellamagate/qwen3-vl-8bSource: LiteLLM data33K$0.15$0.55

Benchmarks

Named effort levels run from lowest to highest. Select a result to see its source details.

Vision

AI2DSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.
CharadesSTASource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.
CharXiv-DSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.
CharXiv-RSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.
DocVQAtestSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.
InfoVQAtestSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.
LVBenchSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.
MMBench-V1.1Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.
MMMU (val)Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.
MMMU-ProSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.
MMStarSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.
MVBenchSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.
ODinWSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.
RealWorldQASource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.
ScreenSpotSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.
Video-MME8 related benchmarks · 43 results
Video-MME
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.
v2
Source: Video-MME-v2
w. sub @ 512, think
w. sub @ 64, instruct
wo sub @ 512, think
wo sub @ 64, instruct
v2 - Avg Acc
Source: Video-MME-v2
w. sub @ 512, think
w. sub @ 64, instruct
wo sub @ 512, think
wo sub @ 64, instruct
v2 - Capability Consistency
Source: Video-MME-v2
w. sub @ 512, think
w. sub @ 64, instruct
wo sub @ 512, think
wo sub @ 64, instruct
v2 - Level 1
Source: Video-MME-v2
w. sub @ 512, think
w. sub @ 64, instruct
wo sub @ 512, think
wo sub @ 64, instruct
v2 - Level 2
Source: Video-MME-v2
w. sub @ 512, think
w. sub @ 64, instruct
wo sub @ 512, think
wo sub @ 64, instruct
Show 12 more resultsHide 12 results
v2 - Level 3
Source: Video-MME-v2
w. sub @ 512, think
w. sub @ 64, instruct
wo sub @ 512, think
wo sub @ 64, instruct
v2 - Reasoning Coherence
Source: Video-MME-v2
w. sub @ 512, think
w. sub @ 64, instruct
wo sub @ 512, think
wo sub @ 64, instruct
VideoMMMUSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.

OCR

CC-OCRSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.
OCR ArenaSource: OCR Arena data · Nov 2026
OCRBenchSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.

Math

AIME 2025Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.
HMMT25Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.
MathVisionSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.
PolyMATHSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.

Knowledge

IncludeSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.
MMLUSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.

Reasoning

ERQASource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.
MuirBenchSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.

Human preference

Arena-Hard v2Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.
MM-MT-BenchSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.

Agentic

BFCL-v3Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.
OSWorldSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.

Coding

Instruction following

IFEvalSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.
Multi-IFSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.

Science

GPQASource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.
SuperGPQASource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.

Factuality

SimpleQASource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.

General

MLVU-MSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.

Other

Nejumi 428 related benchmarks · 28 results
WritingBenchSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it.

Data credits: Data by LLM Stats · Hugging Face official benchmark leaderboards; licence varies by benchmark · Nejumi Leaderboard 4, Weights & Biases Japan; public publisher-selected numeric evaluation results · Physical AI Bench (SHI Labs), MIT (licence) · SnakeBench (Greg Kamradt / ARC Prize Foundation), MIT (licence)

Also known as: llamagate/qwen3-vl-8b, qwen/qwen3-vl-8b-thinking

Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works