# Qwen3 32B

As of October 22, 2026, Qwen3 32B ranks #95 of 123 models overall with 40.8 ±17.4 on the LLMBenchmark Index. Qwen3 32B ranks #63 of 108 in math, its strongest category. Qwen3 32B costs $0.14 per million input tokens and $0.40 per million output tokens. Qwen3 32B was released on April 28, 2025.

Page: https://llmbenchmark.ai/models/qwen3-32b
Data from October 22, 2026. Every figure below names the board it was read from and the day.

- Maker: Qwen
- Released: April 28, 2025
- Price per million input tokens: $0.14
- Price per million output tokens: $0.40
- Context window: 131K tokens
- Overall rank: 95 of 123, by the LLMBenchmark Index
- LLMBenchmark Index: 40.8 ±17.4 (90% margin, from 4 of 7 inputs)
- Artificial Analysis Intelligence Index: 7.3

Benchmark results (45 of 195):

- Aider polyglot: 40.0% (source: Aider polyglot leaderboard, read October 22, 2026)
- AA Intelligence Index: 7.3 (source: Artificial Analysis (via OpenRouter), read October 22, 2026)
- AA Math Index: 73.0 (source: Artificial Analysis, read October 22, 2026)
- GPQA Diamond (Artificial Analysis run): 66.8% (source: Artificial Analysis, read October 22, 2026)
- Humanity's Last Exam (Artificial Analysis run): 7.4% (source: Artificial Analysis, read October 22, 2026)
- MMLU Pro (Artificial Analysis run): 79.8% (source: Artificial Analysis, read October 22, 2026)
- LiveCodeBench (Artificial Analysis run): 54.6% (source: Artificial Analysis, read October 22, 2026)
- SciCode: 36.0% (source: Artificial Analysis, read October 22, 2026)
- Terminal-Bench Hard (Artificial Analysis run): 3.0% (source: Artificial Analysis, read October 22, 2026)
- IFBench: 36.3% (source: Artificial Analysis, read October 22, 2026)
- Long context reasoning (AA-LCR): 0.0% (source: Artificial Analysis, read October 22, 2026)
- τ²-Bench Telecom: 29.8% (source: Artificial Analysis, read October 22, 2026)
- AIME 2025 (Artificial Analysis run): 73.0% (source: Artificial Analysis, read October 22, 2026)
- GPQA Diamond (Epoch AI run): 65.7% (source: Epoch AI benchmarking hub, read October 22, 2026)
- OTIS Mock AIME: 66.9% (source: Epoch AI benchmarking hub, read October 22, 2026)
- DTBench: 67.5% (source: Epoch AI benchmarking hub, read October 22, 2026)
- LMCA: 17.3% (source: Epoch AI benchmarking hub, read October 22, 2026)
- Fiction.LiveBench: 74.2% (source: Epoch AI benchmarking hub, read October 22, 2026)
- Epoch Capabilities Index: 138.4 (source: Epoch AI benchmarking hub, read October 22, 2026)
- LMArena Text: 1340 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: coding: 1357 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: hard prompts: 1334 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: math: 1399 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: instruction following: 1305 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: creative writing: 1297 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: English: 1359 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: non-English: 1317 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: Chinese: 1357 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: German: 1341 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: Russian: 1311 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: multi-turn: 1331 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: longer queries: 1327 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: expert: 1362 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: hard prompts in English: 1350 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: software and IT: 1366 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: mathematical work: 1420 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: science: 1372 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: medicine and healthcare: 1368 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: legal and government: 1324 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: business and finance: 1318 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: writing and language: 1298 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: entertainment and media: 1279 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- Berkeley Function Calling Leaderboard (BFCL): 48.7% (source: Berkeley Function Calling Leaderboard, read October 22, 2026)
- Kagi LLM Benchmark: 48.7% (source: Kagi LLM Benchmarking Project, read October 22, 2026)
- BrowseComp-Plus: 10.7% (source: BrowseComp-Plus, read October 22, 2026)