ai.canshu.com
中文
VRAM Calculator Advisor & TCO API Pricing Compare Models Leaderboard GPUs

Search models, GPUs, and tools

↑↓ Navigate · Enter Open · ★ Watchlist prioritized ai.canshu.com
Alibaba Qwen
Open-Weight Alibaba Qwen · qwen3.6-27b · Text+Image+Video

Qwen3.6 27B Official Specifications, Pricing & Benchmarks

Qwen3.6 27B (qwen3.6-27b) by Alibaba Qwen, released 2026-04-22, featuring 262,144 context window, 65,536 max output tokens, and Text+Image+Video input.

📋 Qwen3.6 27B Official API & System Card Specifications

Verified against official Alibaba Qwen developer documentation & pricing

API Model ID: qwen3.6-27b

1. Identity & Capabilities

API Spec
API Model ID
qwen3.6-27b
Release Date
2026-04-22
Weight License
Open-Weight (API Spec)
Native Reasoning
✅ Supported
Function Calling & JSON Schema
✅ 原生支持
Batch API 异步批处理
✅ Supported (50% off)

2. Context, Output & Modalities

Limits
Max Context Window
262,000 tokens (262K)
Max Output Tokens
65,536 tokens
Input Modalities
Text + Image + Video
Output Modalities
Text
Direct China Access
✅ Direct CN + RMB
Deployment Mode
Open-Weight + Cloud API

3. Official Pricing & Cache

USD / 1M
Input / 1M Tokens
$ 0.6
Cached Input / 1M
$ 0.12 -80%
Output / 1M Tokens
$ 3.6

🎛️ Official Sampling & Reasoning Parameters

Recommended generation settings & API parameter constraints

Official Reasoning & Tool Config
temperature=0.7 top_p=0.95 max_tokens=65,536

📊 Qwen3.6 27B Full Benchmark Scorecard

43 public benchmarks · ranks among 154 models · synced 2026-11-20

⚔️ Compare head-to-head

💻 Coding 8 items

SWE-Bench Verified #24/59 77.2%

A verified subset of 500 software engineering problems from real GitHub issues, validated by human annotators for evaluating language models' ability to resolve real-world coding issues by generating patches for Python codebases.

SWE-bench Multilingual #22/33 71.3%

A multilingual benchmark for issue resolving in software engineering that covers Java, TypeScript, JavaScript, Go, Rust, C, and C++.

Claw-Eval #11/13 60.6%

Claw-Eval tests real-world agentic task completion across complex multi-step scenarios, evaluating a model's ability to use tools, navigate environments, and complete end-to-end tasks autonomously.

Terminal-Bench 2.0 #19/38 59.3%

Terminal-Bench 2.0 is an updated benchmark for testing AI agents' tool use ability to operate a computer via terminal.

SWE-Bench Pro #43/52 53.5%

SWE-Bench Pro is an advanced version of SWE-Bench that evaluates language models on complex, real-world software engineering tasks requiring extended reasoning and multi-step problem solving.

ZClawBench #3/4 53.4%

ZClawBench evaluates Claw-style agent task execution quality, measuring a model's ability to autonomously complete complex multi-step coding tasks in real-world environments.

SkillsBench #6/11 48.2%

SkillsBench evaluates coding agents on self-contained programming tasks, measuring practical engineering skills across diverse software development scenarios.

NL2Repo #21/23 36.2%

NL2Repo evaluates long-horizon coding capabilities including repository-level understanding, where models must generate or modify code across entire repositories from natural language specifications.

📐 Math 11 items

AIME 2026 #10/18 94.1%

All 30 problems from the 2026 American Invitational Mathematics Examination (AIME I and AIME II), testing olympiad-level mathematical reasoning with integer answers from 000-999.

HMMT 2025 #7/11 93.8%

Harvard-MIT Mathematics Tournament 2025 - A prestigious student-organized mathematics competition for high school students featuring two tournaments (November 2025 at MIT and February 2026 at Harvard) with individual tests, team rounds, and guts rounds.

MMLU-Redux #6/15 93.5%

An improved version of the MMLU benchmark featuring manually re-annotated questions to identify and correct errors in the original dataset. Provides more reliable evaluation metrics for language models by addressing dataset quality issues found in the original MMLU.

HMMT25 #3/9 90.7%

Harvard-MIT Mathematics Tournament 2025 - A prestigious student-organized mathematics competition for high school students featuring two tournaments (November 2025 at MIT and February 2026 at Harvard) with individual tests, team rounds, and guts rounds.

MathVista-Mini #3/7 87.4%

MathVista-Mini is a smaller version of the MathVista benchmark that evaluates mathematical reasoning in visual contexts.

MMLU-Pro #11/39 86.2%

A more robust and challenging multi-task language understanding benchmark that extends MMLU by expanding multiple-choice options from 4 to 10, eliminating trivial questions, and focusing on reasoning-intensive tasks.

DynaMath #4/8 85.6%

A multimodal mathematics and reasoning benchmark focused on dynamic visual problem solving.

HMMT Feb 26 #9/11 84.3%

HMMT February 2026 is a math competition benchmark based on problems from the Harvard-MIT Mathematics Tournament, testing advanced mathematical problem-solving and reasoning.

IMO-AnswerBench #16/17 80.8%

IMO-AnswerBench is a benchmark for evaluating mathematical reasoning capabilities on International Mathematical Olympiad (IMO) problems, focusing on answer generation and verification.

SuperGPQA #7/16 66%

SuperGPQA is a comprehensive benchmark that evaluates large language models across 285 graduate-level academic disciplines.

Humanity's Last Exam #49/69 24%

Humanity's Last Exam (HLE) is a multi-modal academic benchmark with 2,500 questions across mathematics, humanities, and natural sciences, designed to test LLM capabilities at the frontier of human knowledge with unambiguous, verifiable solutions.

👁️ Multimodal 20 items

VLMsAreBlind #1/6 97%

A vision-language benchmark that probes blind spots and brittle reasoning in multimodal models.

V* #3/7 94.7%

A visual reasoning benchmark evaluating multimodal inference under challenging spatial and grounded tasks.

MMBench-V1.1 #4/6 92.3%

Version 1.1 of MMBench, an improved bilingual benchmark for assessing multi-modal capabilities of vision-language models through multiple-choice questions in both English and Chinese, providing systematic evaluation across diverse vision-language tasks.

OCRBench #4/5 89.4%

OCRBench: Comprehensive evaluation benchmark for assessing Optical Character Recognition (OCR) capabilities in Large Multimodal Models across text recognition, scene text VQA, and document understanding tasks.

VideoMME w sub. #2/8 87.7%

The first-ever comprehensive evaluation benchmark of Multi-modal LLMs in Video analysis. Features 900 videos (254 hours) with 2,700 question-answer pairs covering 6 primary visual domains and 30 subfields.

MLVU #4/7 86.6%

A comprehensive benchmark for multi-task long video understanding that evaluates multimodal large language models on videos ranging from 3 minutes to 2 hours across 9 distinct tasks including reasoning, captioning, recognition, and summarization.

VideoMMMU #6/14 84.4%

Video-MMMU evaluates Large Multimodal Models' ability to acquire knowledge from expert-level professional videos across six disciplines through three cognitive stages: perception, comprehension, and adaptation.

RealWorldQA #9/13 84.1%

RealWorldQA is a benchmark designed to evaluate basic real-world spatial understanding capabilities of multimodal models.

MMMU #4/16 82.9%

MMMU (Massive Multi-discipline Multimodal Understanding) is a benchmark designed to evaluate multimodal models on college-level subject knowledge and deliberate reasoning.

MMStar #4/8 81.4%

MMStar is an elite vision-indispensable multimodal benchmark comprising 1,500 challenge samples meticulously selected by humans to evaluate 6 core capabilities and 18 detailed axes.

CC-OCR #4/6 81.2%

A comprehensive OCR benchmark for evaluating Large Multimodal Models (LMMs) in literacy. Comprises four OCR-centric tracks: multi-scene text reading, multilingual text reading, document parsing, and key information extraction.

CharXiv-R #27/41 78.4%

CharXiv-R is the reasoning component of the CharXiv benchmark, focusing on complex reasoning questions that require synthesizing information across visual chart elements.

MMMU-Pro #25/44 75.8%

A more robust multi-discipline multimodal understanding benchmark that enhances MMMU through a three-step process: filtering text-only answerable questions, augmenting candidate options, and introducing vision-only input settings.

MVBench #3/7 75.5%

A comprehensive multi-modal video understanding benchmark covering 20 challenging video tasks that require temporal understanding beyond single-frame analysis.

ERQA #9/13 62.5%

Embodied Reasoning Question Answering benchmark consisting of 400 multiple-choice visual questions across spatial reasoning, trajectory reasoning, action reasoning, state estimation, and multi-view reasoning for evaluating AI capabilities in physical world interactions.

CountBench #2/7 1

CountBench evaluates object counting capabilities in visual understanding.

RefCOCO-avg #2/6 0.9

RefCOCO-avg measures object grounding accuracy averaged across RefCOCO, RefCOCO+, and RefCOCOg benchmarks.

EmbSpatialBench #1/6 0.8

EmbSpatialBench evaluates embodied spatial understanding and reasoning capabilities.

RefSpatialBench #2/7 0.7

RefSpatialBench evaluates spatial reference understanding and grounding.

SimpleVQA #12/13 0.6

SimpleVQA is a visual question answering benchmark focused on simple queries.

Source: ZeroEval / llm-stats.com (✓ = third-party verified, others vendor-reported).

Closest Self-Hostable Open-Weight Alternatives to Qwen3.6 27B

Open-weight models with full HuggingFace weights for private GPU / Mac deployment

Compare with these →

Frequently Asked Questions

Is Qwen3.6 27B open-source or proprietary?

Qwen3.6 27B (qwen3.6-27b) is a proprietary commercial model by Alibaba Qwen. Weights and internal layer dimensions are undisclosed; it is accessed via cloud API.

What are the context limit and API pricing of Qwen3.6 27B?

Qwen3.6 27B supports 262K input context and 65,536 max output tokens, priced at $0.6/1M input ($0.12/1M cached) and $3.6/1M output.