llmboard.ai
Benchmarks
CompareRankings
llmboard.ai
Benchmarks
CompareRankings
HomeModelsQwen3.6 27B

Alibaba Cloud / Qwen Team model product

Qwen3.6 27B

Qwen3.6-27B is a dense 27-billion-parameter multimodal model in the Qwen3.6 series, supporting both vision-language thinking and non-thinking modes in a single unified checkpoint. The 64-layer language model uses a hybrid layout of 16 repeats of (3 × Gated DeltaNet → FFN, 1 × Gated Attention → FFN) with hidden dim 5120 and FFN intermediate 17408 — Gated DeltaNet has 48/16 heads for V/QK (head dim 128) and Gated Attention has 24/4 heads for Q/KV (head dim 256). It supports a native 262,144-token context extensible to ~1,010,000 via YaRN and is trained with multi-token prediction. The release delivers flagship-level agentic coding, surpassing the previous-generation open-source flagship Qwen3.5-397B-A17B (397B total / 17B active) on every major coding benchmark including SWE-bench Verified (77.2), SWE-bench Pro (53.5), Terminal-Bench 2.0 (59.3), and SkillsBench (48.2), and reaches 87.8 on GPQA Diamond. Released as open weights under Apache 2.0; accessible via Qwen Studio with the Alibaba Cloud Model Studio API coming soon.

Updated Aug 10, 2026. Default version: Qwen3.6-27B

Compare
LLMBoard score72.2Qwen3.6-27B
Coverage80%8 benchmark families
Context window262.1KTokens
Official input priceN/AOfficial price unavailable

On this page

  • Specification
  • Capability
  • Benchmarks
  • Arena
  • Pricing
  • Versions
  • About
  • Compare
  • Similar models
  • FAQ

Model specification

Structured fields from the published default version.

Version
Qwen3.6-27B
Released
Apr 21, 2026
Knowledge cutoff
Unknown
Parameters
27.8B
Context window
262.1K
Max output
65.5K
Inputs
audio, image, text, video
Outputs
text
Open weights
Yes
License
Apache 2.0

Capability profile

This profile uses the latest version under this unique model that has a calculated LLMBoard score. Arena and price are excluded.

Qwen3.6-27B category scores

Benchmark results

Published benchmark records for the scored version Qwen3.6-27B.

44 rows
Columns

Show columns

EmbSpatialBench0.818100.0%CAug 7, 2026
RefSpatialBench0.716100.0%CAug 7, 2026
MVBench0.821793.8%CAug 7, 2026
QwenWebBench1487.0220.0%CAug 7, 2026
RefCOCO-avg0.92783.3%CAug 7, 2026
VideoMME w sub.0.921088.9%CAug 7, 2026
VLMsAreBlind1.02466.7%CAug 7, 2026
CountBench1.03660.0%CAug 7, 2026
V*0.93766.7%CAug 7, 2026
ZClawBench0.53433.3%CAug 7, 2026
AndroidWorld0.7440.0%CAug 7, 2026
DynaMath0.94750.0%CAug 7, 2026
HMMT250.942587.5%CAug 7, 2026
MathVista-Mini0.942386.4%CAug 7, 2026
MLVU0.941066.7%CAug 7, 2026
MMBench-V1.10.941882.3%CAug 7, 2026
MMStar0.842285.7%CAug 7, 2026
SkillsBench0.55733.3%CAug 7, 2026
CC-OCR0.861870.6%CAug 7, 2026
C-Eval0.961870.6%CAug 7, 2026
AIME 20260.971762.5%CAug 7, 2026
MMMU0.886388.7%CAug 7, 2026
SuperGPQA0.783478.8%CAug 7, 2026
HMMT Feb 260.891120.0%CAug 7, 2026
MMLU-Redux0.994883.0%CAug 7, 2026
OCRBench0.992261.9%CAug 7, 2026
RealWorldQA0.892668.0%CAug 7, 2026
VideoMMMU0.892668.0%CAug 7, 2026
Claw-Eval0.6101325.0%CAug 7, 2026
ERQA0.6102359.1%CAug 7, 2026
HMMT 20250.9113368.8%CAug 7, 2026
LiveCodeBench v60.8115380.8%CAug 7, 2026
MMLU-Pro0.91212991.4%CAug 7, 2026
SimpleVQA0.612138.3%CAug 7, 2026
NL2Repo0.413147.7%CAug 7, 2026
IMO-AnswerBench0.8151922.2%CAug 7, 2026
SWE-bench Multilingual0.7173451.5%CAug 7, 2026
Terminal-Bench 2.00.6224956.3%CAug 7, 2026
CharXiv-R0.8234752.2%CAug 7, 2026
SWE-Bench Verified0.82610475.7%CAug 7, 2026
MMMU-Pro0.8306554.7%CAug 7, 2026
SWE-Bench Pro0.5364418.6%CAug 7, 2026
GPQA0.93723384.5%CAug 7, 2026
Humanity's Last Exam0.2529244.0%CAug 7, 2026

Arena results

Preference and agent-evaluation signals from published Arena datasets.

No published Arena match

The default version has no published Arena rows, or its source alias has not been resolved.

Pricing

Official vendor API PAYG pricing is summarized first. The table then lists individual provider offerings without treating their minimum as the official price.

Official API
N/A
Official provider
N/A
Lowest third-party
From $0.20 input, $1.5 output per 1M via CrofAI
Tracked offerings
5
4 rows
Columns

Show columns

QVACqwen3.6-27bglobalN/AN/A262.1KAug 7, 2026
CrofAIqwen3.6-27bglobal$0.20$1.5262.1KAug 7, 2026
Cortecsqwen3.6-27bglobal$0.446$3.01262KAug 7, 2026
OVHcloud AI Endpointsqwen3.6-27bglobal$0.47$3.19262.1KAug 7, 2026

Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.

Qwen3.6 27B versions

All published versions linked to this unique model. The score columns identify the version used by the current overall ranking.

1 rows
Columns

Show columns

Qwen3.6-27BApr 21, 202672.227.8B262.1K65.5KYesApache 2.0

What is Qwen3.6 27B?

A concise description based on the published model registry.

Qwen3.6-27B is a dense 27-billion-parameter multimodal model in the Qwen3.6 series, supporting both vision-language thinking and non-thinking modes in a single unified checkpoint. The 64-layer language model uses a hybrid layout of 16 repeats of (3 × Gated DeltaNet → FFN, 1 × Gated Attention → FFN) with hidden dim 5120 and FFN intermediate 17408 — Gated DeltaNet has 48/16 heads for V/QK (head dim 128) and Gated Attention has 24/4 heads for Q/KV (head dim 256). It supports a native 262,144-token context extensible to ~1,010,000 via YaRN and is trained with multi-token prediction. The release delivers flagship-level agentic coding, surpassing the previous-generation open-source flagship Qwen3.5-397B-A17B (397B total / 17B active) on every major coding benchmark including SWE-bench Verified (77.2), SWE-bench Pro (53.5), Terminal-Bench 2.0 (59.3), and SkillsBench (48.2), and reaches 87.8 on GPQA Diamond. Released as open weights under Apache 2.0; accessible via Qwen Studio with the Alibaba Cloud Model Studio API coming soon.

Use the benchmark, Arena and pricing sections above as separate evidence. A missing field means the current data snapshot does not support that claim.

Data snapshot: 2026-08-07. Editorial model content is not available in the backend.

Qwen3.6 27B vs nearby models

Open a comparison with the three ranked models immediately above and below this model.

Qwen3.6 27BvsQwen3.7 PlusQwen3.6 27BvsGemini 3 FlashQwen3.6 27BvsQwen3.6 PlusQwen3.6 27BvsNemotron 3 UltraQwen3.6 27BvsDeepSeek-V4-FlashQwen3.6 27BvsKimi K2.5

Models similar to Qwen3.6 27B

Recommendations prioritize the same model type and family, then the closest published LLMBoard score.

#25+0.2
AC

Qwen3.6 Plus

Alibaba Cloud / Qwen Team

72.4 LLMBoard

DetailsCompare
#23+0.9
AC

Qwen3.7 Plus

Alibaba Cloud / Qwen Team

73.1 LLMBoard

DetailsCompare
#19+3.0
AC

Qwen3.5 397B A17B

Alibaba Cloud / Qwen Team

75.2 LLMBoard

DetailsCompare
#39-6.5
AC

Qwen3 235B A22B Thinking

Alibaba Cloud / Qwen Team

65.7 LLMBoard

DetailsCompare
#50-10.8
AC

Qwen3.5 122B A10B

Alibaba Cloud / Qwen Team

61.4 LLMBoard

DetailsCompare
#57-13.7
AC

Qwen3 VL 235B A22B Thinking

Alibaba Cloud / Qwen Team

58.5 LLMBoard

DetailsCompare

FAQ

Common questions about Qwen3.6 27B.

When was Qwen3.6 27B released?

Qwen3.6 27B's default version was released on Apr 21, 2026.

How much does Qwen3.6 27B cost?

No official standard PAYG price is currently available for Qwen3.6 27B. The lowest tracked third-party offer starts at $0.20 input and $1.5 output via CrofAI.

Who created Qwen3.6 27B?

Qwen3.6 27B is published under Alibaba Cloud / Qwen Team in the model registry.

What is the context window for Qwen3.6 27B?

The default version has a 262.1K token context window.

Is Qwen3.6 27B open weight?

Yes. The default version is marked as open weight under Apache 2.0.

How many API providers offer Qwen3.6 27B?

5 published provider offerings are linked to the default version.

What models should I compare Qwen3.6 27B with?

Nearby ranked alternatives include Qwen3.7 Plus, Gemini 3 Flash, Qwen3.6 Plus.

Rankings

OverallCodingText ArenaPricing

Modalities

Image GenerationVideo GenerationSpeech-to-TextEmbeddings

Benchmarks

All BenchmarksReasoningMathCoding

Vendors

All VendorsOpenAIAnthropicGoogle
llmboard.aiCopyright 2026 llmboard.ai