Skip to content

Alibaba · Chat

Qwen3 32B

Alibaba 🇨🇳ChatQwen familyOpen weights. One of its sources says the weights are publicly available.The maker lists this model as deprecated (16 Feb 2026).
From Aiml API's listing:
Alibaba's Qwen 3 32B is a multilingual model with instruction-following capabilities and enhanced reasoning.
llminfo IQ
115.6
Maker's price
$1.225 blended / 1M
Max context
131K
Listings
65

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/Qwen3-32Bllminfo IQ 115.6, ranked among chat models. From 234 benchmarks. Plenty of data behind it.How llminfo IQ is computedCoding IQ 84.8, 95% interval 76–93. From 9 benchmark families across 8 sources. A fair amount 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 (35.8). Costs more per task than its benchmark results predict. From 7 benchmarks with measured task costs. A fair amount of data behind it.How Value for money worksReleased 29 Apr 2025 Training-data cutoff, as its sources state it Maker HQ: CN (source)
All rankings: #223 on NanoGPT recentlySources: LMArena data, DeepInfra data, LiteLLM data, Models.dev data, AI/ML API data, OpenRouter data docs, llminfo researchPrimary data from LMArena from DeepInfra: name from LiteLLM: deprecation date from Models.dev: weights availability from AI/ML API: primary mode, description from OpenRouter: knowledge cutoff from llminfo research: release date
Tags from its sources:instructtoolsopen-sourcereasoningmultilingualtool-calling

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 8 peers that say either way have open weights.

Trade-offs

Deprecated by its maker

The maker has already deprecated this model (16 Feb 2026).

We only make a claim when there's enough data on models of the same type. Updated 22 Nov, 04:34 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
Text

Capabilities

Supported operations and provider-stated features

ReasoningFunction callingStructured outputPrompt cachingBatch API

Listings

 Alibaba Cloudqwen3-32bSource: Models.dev data131K$0.70$2.80
 Lambda AIlambda_ai/qwen3-32b-fp8 · fp8Source: LiteLLM data131K$0.05$0.10
 OpenRouterInference by DeepInfraqwen/qwen3-32b · fp8/flex · fp8Source: OpenRouter data docs41K$0.064$0.224
 Fastrouterqwen/qwen3-32bSource: FastRouter data41K$0.08$0.28
 FlowdotQwen/Qwen3-32BSource: FlowDot data41K$0.08$0.28
 Hugging FaceInference by DeepInfraQwen/Qwen3-32BSource: Hugging Face Inference Providers data41K$0.08$0.28
 Kiloqwen/qwen3-32bSources: Kilo AI Gateway data, Models.dev data41K$0.08$0.28
 NanoGPTqwen/qwen3-32bSources: Models.dev data, NanoGPT41K$0.08$0.28
 Nous Portalqwen/qwen3-32bSource: Nous Portal Inference Gateway data41K$0.08$0.28
 AbacusQwen/Qwen3-32BSource: Models.dev data131K$0.08$0.28

8 listings without a price are hidden.

Benchmarks

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

Multilingual

IndicParam42 related benchmarks · 42 results
Show 36 more resultsHide 36 results

Knowledge

Human preference

Arena ELO24 related benchmarks · 26 results
Arena ELO
Source: LMArena data · Nov 2026
Arena Hard
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Arena-Hard-v2.0-Preview
Source: Arena-Hard-Auto data
Creative Writing, Ensemble GPT-4.1 and Gemini 2.5 as Judges (Best Configuration for Creative Writing)
Hard Prompt, Style Control, and Gemini-2.5 as Judge (Official Configuration)
Hard Prompt, Style Control, and GPT-4.1 as Judge (If prefer OpenAI API)
Chinese
Source: LMArena data · Nov 2026
Creative Writing
Source: LMArena data · Nov 2026
English
Source: LMArena data · Nov 2026
Show 18 more resultsHide 18 results
Exclude Ties
Source: LMArena data · Nov 2026
Expert
Source: LMArena data · Nov 2026
German
Source: LMArena data · Nov 2026
Hard Prompts
Source: LMArena data · Nov 2026
Hard Prompts English
Source: LMArena data · Nov 2026
Industry Mathematical
Source: LMArena data · Nov 2026
Longer Query
Source: LMArena data · Nov 2026
Multi Turn
Source: LMArena data · Nov 2026
Non English
Source: LMArena data · Nov 2026
Russian
Source: LMArena data · Nov 2026
Style-Controlled
Source: LMArena data · Nov 2026

Agentic

BFCL8 related benchmarks · 8 results
BFCL
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Show 2 more resultsHide 2 results
GAIA4 results
Source: GAIA Leaderboard data · Jun 2025
Difficulty level
Difficulty not reported
Source: OpenRouter benchmark exports
qwen3-32b-04-28 · 90-day results
Across 3 providers lowest 44.9% median 48.3% highest 48.7%OpenRouter's automatic routing 46%
Show results from 3 providers and OpenRouter's automatic routingHide results from 3 providers and OpenRouter's automatic routing
  • Nebius
  • DeepInfra
  • OpenRouter's automatic routing
  • Tenstorrent

Science

GPQA Diamond7 results
Source: Epoch AI Benchmarking Hub data
Reasoning effort
Effort not reported
Source: OpenRouter benchmark exports
qwen3-32b-04-28 · 90-day results
Across 4 providers lowest 54.7% median 58.9% highest 65%OpenRouter's automatic routing 59.9%
Show results from 4 providers and OpenRouter's automatic routingHide results from 4 providers and OpenRouter's automatic routing
  • Groq
  • Nebius
  • OpenRouter's automatic routing
  • DeepInfra
  • Tenstorrent

Coding

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

Long context

RULERSource: RULER official results data · Reported by the model's developer. The source relaying it did not verify it.

General

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

Vision

Math

AIME 2024Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
AIME 2025Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
LiveBenchSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim

Instruction following

Reasoning

Factuality

Other

Horangi36 related benchmarks · 36 results
Nejumi 428 related benchmarks · 28 results
ALT_バイアス
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled
ALT_倫理・道徳
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled
ALT_制御性
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled
ALT_堅牢性
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled
ALT_毒性
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled
ALT_真実性
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled
Show 22 more resultsHide 22 results
AVG_jaster_0shot
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled
AVG_jaster_2shots
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled
GLP_アプリケーション開発
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled
GLP_コーディング
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled
GLP_一般的知識
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled
GLP_基礎的言語性能
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled
GLP_専門的知識
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled
GLP_応用的言語性能
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled
GLP_情報検索
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled
GLP_意味解析
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled
GLP_抽象的推論
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled
GLP_推論能力
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled
GLP_数学的推論
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled
GLP_構文解析
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled
GLP_知識・質問応答
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled
GLP_翻訳
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled
GLP_表現
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled
GLP_論理的推論
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled
GLP_関数呼び出し
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled
TOTAL_SCORE
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled
アラインメント(ALT)_AVG
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled
汎用的言語性能(GLP)_AVG
Source: Nejumi Leaderboard 4 data docs
Qwen/Qwen3-32B: reasoning-enabled

Data credits: Data by LLM Stats · Aider LLM Leaderboards (Aider-AI), Apache-2.0 (licence) · Epoch AI Benchmarking Hub, CC BY 4.0 (licence) · SEA-HELM leaderboard, AI Singapore (AI Products team), National University of Singapore. Published aggregate scores; results licence not stated (SEA-HELM code is MIT). · LMArena, CC BY 4.0 (licence) · Arena-Hard-Auto, LMArena, Apache-2.0 (licence) · Berkeley Function-Calling Leaderboard (Gorilla, UC Berkeley), Apache-2.0 (licence) · EQ-Bench (Samuel J. Paech) — cite paech2023eqbench · OpenRouter public benchmark leaderboard and search-lane exports, CC BY 4.0. Fractional source scores converted to percentages; immutable dated snapshots. (licence) · Horangi 4, Weights & Biases Korea; public publisher-selected numeric evaluation results · JMLE2026-Bench and IgakuQA119 by Naoto Iwase; results licence not stated (code MIT, question data CC BY 4.0). · Maheshwari et al., IndicParam, arXiv:2512.00333v3. Cited numeric scores under the owner's 2026-10-10 task-only restriction scope decision; no results licence established. Tasks, prompts and generations excluded. · Hugging Face official benchmark leaderboards; licence varies by benchmark · LiveMCPBench leaderboard, ICIP-CAS (Mo et al., 2025). Sheet linked from the Apache-2.0 project README; sheet licence not separately stated. · LLM Chess by Maxim Saplin, Apache-2.0 (Copyright [2024] [Maxim Saplin]) (licence) · MMLU-ProX: Xuan et al., arXiv:2503.10497. Public project-page results; no results licence stated. · Nejumi Leaderboard 4, Weights & Biases Japan; public publisher-selected numeric evaluation results · Open Arabic LLM Leaderboard (OALL), Apache-2.0 (licence) · pfgen-bench, Preferred Networks. Apache-2.0 except LLM-generated outputs; aggregate scores only. (licence) · NVIDIA RULER, Hsieh et al. Official README numeric table; no separate results licence asserted. · SnakeBench (Greg Kamradt / ARC Prize Foundation), MIT (licence) · Swallow LLM Leaderboard v2, Swallow LLM team (Institute of Science Tokyo). Contents CC BY 4.0; Japanese post-trained task scores on the source's 0..1 scale. (licence) · SWE-Effi (Centre for Software Excellence), Apache-2.0; resolved rate converted to percent; resource figures are per-task means (licence) · UzLiB, Tahrirchi (Abror Shopulatov). Public LEADERBOARD.md numbers; repository MIT covers software and documentation, no separate results licence stated and none asserted. · Vectara Hallucination Leaderboard, Apache-2.0 (licence) · WebWalkerQA (Alibaba NLP), Apache-2.0 (licence)

Also known as: alibaba/qwen-3-32b, alibaba/qwen3-32b, bedrock/ap-northeast-1/qwen.qwen3-32b-v1:0, bedrock/ap-south-1/qwen.qwen3-32b-v1:0, bedrock/ap-southeast-3/qwen.qwen3-32b-v1:0, bedrock/eu-central-1/qwen.qwen3-32b-v1:0, bedrock/eu-south-1/qwen.qwen3-32b-v1:0, bedrock/eu-west-1/qwen.qwen3-32b-v1:0, bedrock/eu-west-2/qwen.qwen3-32b-v1:0, bedrock/sa-east-1/qwen.qwen3-32b-v1:0, bedrock_mantle/ap-northeast-1/qwen.qwen3-32b, bedrock_mantle/ap-south-1/qwen.qwen3-32b, bedrock_mantle/ap-southeast-2/qwen.qwen3-32b, bedrock_mantle/ap-southeast-3/qwen.qwen3-32b, bedrock_mantle/eu-central-1/qwen.qwen3-32b, bedrock_mantle/eu-south-1/qwen.qwen3-32b, bedrock_mantle/eu-west-1/qwen.qwen3-32b, bedrock_mantle/eu-west-2/qwen.qwen3-32b, bedrock_mantle/qwen.qwen3-32b, bedrock_mantle/sa-east-1/qwen.qwen3-32b, cerebras/qwen-3-32b, deepinfra/qwen/qwen3-32b, fireworks_ai/accounts/fireworks/models/qwen3-32b, gmi/qwen/qwen3-32b-fp8, gradient_ai/alibaba-qwen3-32b, lambda_ai/qwen3-32b-fp8, nebius/qwen/qwen3-32b, novita/qwen/qwen3-32b-fp8, ovhcloud/qwen3-32b, qwen-3-32b, qwen.qwen3-32b-v1:0, qwen/qwen3-32b, qwen/qwen3-32b-fp8, qwen/qwen3-32b-v1:0, scx-ai/qwen3-32b, sophnet/qwen3-32b, vercel_ai_gateway/alibaba/qwen-3-32b

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