Skip to content

Alibaba · Chat

Qwen3.6 35B A3B

Alibaba 🇨🇳ChatQwen familyOpen weights. One of its sources says the weights are publicly available.
From Empiriolabs's listing:
Qwen3.6 35B A3B is a 256-expert mixture-of-experts reasoning model with 128K context, function tools, and strict structured JSON output.
llminfo IQ
119.8
Max context
1M
Listings
117

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/Qwen3.6-35B-A3B-FP8 Model architecture, as its sources state it llminfo IQ 119.8, ranked among chat models. From 160 benchmarks. Plenty of data behind it.How llminfo IQ is computedCoding IQ 94.8, 95% interval 81–109. From 11 benchmark families across 4 sources. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.How Coding IQ is computedPoor value (38.5). Costs more per task than its benchmark results predict. From 8 benchmarks with measured task costs. Plenty of data behind it.How Value for money worksReleased 15 Apr 2026Maker HQ: CN (source)
All rankings: #36 on OpenRouter today #58 on OpenRouter this week #69 on OpenRouter this month #82 on NanoGPT recentlySources: Alibaba Cloud Model Studio (DashScope) Multi-Region API, Ambient data, EmpirioLabs data, flex.ai API data, Models.dev data, Morpheus Inference Marketplace data, Hugging FacePrimary data from Alibaba Cloud Model Studio (DashScope) Multi-Region API from Ambient: name from EmpirioLabs: primary mode, description from flex.ai API: hugging face id from Models.dev: weights availability from Morpheus Inference Marketplace: architecture from Hugging Face: release date
Tags from its sources:instructreasoningtoolsvision

Advantages and trade-offs

Compared with 10 chat models, picked mainly by popularity and IQ

Advantages

Open weights available

The weights are public. 3 of the 7 peers that say either way have open weights.

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 8 maker's prices; 2 market median prices.

Input & output

as stated by the model's providers
Text
Images
Video
Text

Capabilities

Supported operations and provider-stated features

VisionReasoningFunction callingStructured outputPrompt cachingWeb search

Listings

 Alibaba CloudRegions: 🇸🇬🇨🇳+2 ap-southeast-1, cn-beijing, eu-central-1, us-east-1qwen3.6-35b-a3bSources: Alibaba Cloud Model Studio (DashScope) Multi-Region API, Alibaba Cloud Qwen, Models.dev data262K$0.375$2.25——
 OpenRouterInference by DeepInfraqwen/qwen3.6-35b-a3b · fp8/flex · fp8Source: OpenRouter data docs262K$0.08$0.76191,981.5ms
 Salad Cloudqwen3.6-35b-a3bSource: Models.dev data262K$0.09$0.60——
 Fastrouterqwen/qwen3.6-35b-a3bSource: FastRouter data262K$0.10$0.95——
 FlowdotQwen/Qwen3.6-35B-A3BSource: FlowDot data262K$0.10$0.95——
 Hugging FaceInference by AkashQwen/Qwen3.6-35B-A3BSource: Hugging Face Inference Providers data262K$0.10$0.90——
 Kiloqwen/qwen3.6-35b-a3bSources: Kilo AI Gateway data, Models.dev data262K$0.10cache read $0.025$0.90——
 Nous Portalqwen/qwen3.6-35b-a3bSource: Nous Portal Inference Gateway data262K$0.10cache read $0.025$0.90——
 Wavespeedqwen/qwen3.6-35b-a3bSource: WaveSpeed AI data262K$0.10cache read $0.05$0.90——
 AkashQwen/Qwen3.6-35B-A3B · fp8Source: Akash data262K$0.10$0.90——
Show all 102 listings

Showing 10 of 102: the maker's own listing first, then the cheapest per 1M input tokens from each provider. 9 listings with suspiciously low prices are in the full list only.

Also sold by 302ai, Aihubmix, Ambient, Atlascloud, Cortecs AI, Darkbloom, DeepInfra, Doubleword, Empiriolabs, Engy, Evroc, Featherless AI, Flexai, Gmicloud, Greenpt, Ionet, Klavo, Libertai, Llmgateway, Llmgateway Providers, Mallard, Meganova, Merge Gateway, NanoGPT, Near AI, Netmind, Novita AI, Openroute Cn, Opper, Orcarouter, Parasail, Phala, Pinstripes, Ppio, Redpill, Scaleway, SiliconFlow, Siliconflow Cn, Streamlake, Thinking Machines, Venice, Vercel, Weights & Biases, Zerogpu.

15 listings without a price are hidden.

These providers advertise this model. Researchers have found some resellers serving a different, cheaper model, so see how to check whether a provider meets your needs.

Included in plans

Umans Code ProumansSubscription$20/moslows down past the allowancePlan page
Umans Code MaxumansSubscription$50/moslows down past the allowancePlan page

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. claim
LVBenchSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
MLVUSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
MMMUSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
MVBenchSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
ODinWSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
SimpleVQASource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
VideoMMMUSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim

Agentic

GAIA4 results
Source: GAIA Leaderboard data · Jun 2026
Difficulty level
Difficulty not reported
MCP AtlasSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
MCP-MarkSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Source: OpenRouter benchmark exports
15 Apr 2026 version · 90-day results
Across 11 providers lowest 0.762% median 74.4% highest 76.2%OpenRouter's automatic routing 74.9%
Show results from 11 providers and OpenRouter's automatic routingHide results from 11 providers and OpenRouter's automatic routing
  • DekaLLM
  • Phala
  • Darkbloom
  • OpenRouter's automatic routing
  • Parasail
  • DeepInfra
  • Reka
  • AkashML
  • PrimeIntellect
  • AtlasCloud
  • Io Net
  • Venice

Science

GPQASource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
GPQA Diamond16 results
Source: Epoch AI Benchmarking Hub data
Reasoning effort
Effort not reported
Source: OpenRouter benchmark exports
15 Apr 2026 version · 90-day results
Across 13 providers lowest 69.6% median 82.9% highest 85.3%OpenRouter's automatic routing 80%
Show results from 13 providers and OpenRouter's automatic routingHide results from 13 providers and OpenRouter's automatic routing
  • Parasail
  • AtlasCloud
  • SiliconFlow
  • AkashML
  • Reka
  • Venice
  • CoreWeave
  • Phala
  • DekaLLM
  • OpenRouter's automatic routing
  • PrimeIntellect
  • DeepInfra
  • Darkbloom
  • Io Net
SuperGPQASource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim

Coding

NL2RepoSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
SWE-bench Verified6 related benchmarks · 6 results
SWE-bench Verified
Source: Hugging Face Official Benchmarks data · Reported by the model's developer. The source relaying it did not verify it. claim
SWE-bench Multilingual
Source: Hugging Face Official Benchmarks data · Reported by the model's developer. The source relaying it did not verify it. claim

Long context

Math

HMMT25Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim

Reasoning

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

Multilingual

General

Human preference

Safety

Knowledge

OCR

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

Other

Nejumi 428 related benchmarks · 28 results
ALT_バイアス
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
ALT_倫理・道徳
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
ALT_制御性
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
ALT_堅牢性
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
ALT_毒性
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
ALT_真実性
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
Show 22 more resultsHide 22 results
AVG_jaster_0shot
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
AVG_jaster_2shots
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
GLP_アプリケーション開発
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
GLP_コーディング
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
GLP_一般的知識
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
GLP_基礎的言語性能
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
GLP_専門的知識
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
GLP_応用的言語性能
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
GLP_情報検索
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
GLP_意味解析
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
GLP_抽象的推論
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
GLP_推論能力
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
GLP_数学的推論
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
GLP_構文解析
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
GLP_知識・質問応答
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
GLP_翻訳
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
GLP_表現
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
GLP_論理的推論
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
GLP_関数呼び出し
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
TOTAL_SCORE
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
アラインメント(ALT)_AVG
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
汎用的言語性能(GLP)_AVG
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3.6-35B-A3B: reasoning-enabled
Sayf16 related benchmarks · 16 results
Show 10 more resultsHide 10 results

Data credits: Data by LLM Stats · Hugging Face official benchmark leaderboards; licence varies by benchmark · SEA-HELM leaderboard, AI Singapore (AI Products team), National University of Singapore. Published aggregate scores; results licence not stated (SEA-HELM code is MIT). · Baromètre de l’écriture IA, Johannes Eva; public numeric ranking data, © 2026 Johannes Eva, no results licence stated · Epoch AI Benchmarking Hub, CC BY 4.0 (licence) · OpenRouter public benchmark leaderboard and search-lane exports, CC BY 4.0. Fractional source scores converted to percentages; immutable dated snapshots. (licence) · EQ-Bench (Samuel J. Paech) — cite paech2023eqbench · Nejumi Leaderboard 4, Weights & Biases Japan; public publisher-selected numeric evaluation results · Sayf-eval. Copyright (c) 2026 Qatar Computing Research Institute (QCRI). MIT License. (licence) · SnakeBench (Greg Kamradt / ARC Prize Foundation), MIT (licence)

Also known as: aihubmix/qwen3.6-35b-a3b, alibaba/qwen3.6-35b-a3b, deepinfra/qwen/qwen3.6-35b-a3b, e2ee-qwen3-6-35b-a3b, empiriolabs/qwen3-6-35b-a3b, gmi/qwen/qwen3.6-35b-a3b, libertai/qwen3.6-35b-a3b, libertai/qwen3.6-35b-a3b-thinking, novita/qwen/qwen3.6-35b-a3b, novita/qwen3.6-35b-a3b, parasail-qwen3p6-35b-a3b, pinstripes/ps/qwen3.6-35b-a3b, qwen 3.6 35b a3b, qwen 3.6 35b a3b fp8, qwen-3.6-35b-a3b, qwen/qwen3.6-35b-a3b, qwen/qwen3.6-35b-a3b-fp8, qwen/qwen3.6-35b-a3b:thinking, qwen3-6-35b-a3b, qwen3.6-35b-a3b-fp8, qwen3.6-35b-a3b-vl-mtp-mxfp8, scaleway/qwen/qwen3.6-35b-a3b, streamlake/qwen3.6-35b-a3b, tee/qwen3.6-35b-a3b, wandb/qwen/qwen3.6-35b-a3b

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