Leaderboard / Alibaba (Qwen)

Qwen3 32B

Name: Qwen3 32B
Price: 0.13 USD
Author: Alibaba (Qwen)

Alibaba (Qwen) · Qwen3 · released 2025-04-29 · data checked 2026-11-01 Needs verification

Qwen3 32B is a 32.8B-parameter dense model from Alibaba (Qwen) (China), released 2025-04-29 under the Apache-2.0 license. It scores 52.4 on the Small LM Index, #26 of 34 medium models. It needs about 23.2 GB of memory at 4-bit with an 8K-token context and costs $0.13 per million tokens through the cheapest API provider.

Small LM Index

52.4

#26 of 34 scored medium models

Likely 47.3–57.3 · from 2 of 6 core benchmarks

Parameters

32.8B

Dense

Price per 1M tokens

$0.13

$0.080 in / $0.28 out · blended 3:1

Context window

131K

tokens

Benchmarks

Each bar shows this model's score. The tick marks the best medium model on that benchmark. Sources: 2 reported by Alibaba (Qwen). How we label sources.

AIME Lab-reported Math · AIME 2025 (thinking)72.9

Class best: Ternary Bonsai 2 27B 95.8

LiveCodeBench Lab-reported Coding · LiveCodeBench v565.7

Class best: Qwen3.8 27B 90.3

Not reported: estimated

What the index model expects from this model's other scores. Not a measurement.

MMLU-Pro≈ 74

GPQA Diamond≈ 66

Humanity's Last Exam≈ 7

SWE-bench Verified≈ 34

Specifications

Lab: Alibaba (Qwen) (CN · China)
Architecture: Dense transformer
Total parameters: 32.8B
Active parameters: All (dense)
Size class: Medium
Context window: 131K tokens
Input modalities: text
Reasoning: Hybrid (thinking on/off)
License: Apache-2.0
Openness: Open weights
Memory to run: ~23.2 GB at 4-bit, 8K context (more sizes)
Hugging Face: 3M downloads (30 days) · 791 likes
API providers: 2 on OpenRouter
API price: $0.080 in / $0.28 out per 1M (OpenRouter, 2026-11-03)

Weights on Hugging Face ↗Try on OpenRouter ↗

Compare with

Qwen3 32B vs Nanbeige4.3 33B (33.1B, index 77.8)
Qwen3 32B vs Sarvam 30B (32.2B, index 61.6)
Qwen3 32B vs OLMo 3.1 32B Think (32.2B, index 57.0)
Qwen3 32B vs DeepSeek-R1-Distill-Qwen-32B (32.8B, index 50.9)

Sources

https://arxiv.org/abs/2505.09388

Something wrong here? Report it.

Run it locally

Uses Qwen/Qwen3-32B-GGUF (Q4_K_M). Swap the tag for another size: Q4_K_M, Q5_0, Q5_K_M, Q6_K, Q8_0.

Ollama

ollama run hf.co/Qwen/Qwen3-32B-GGUF:Q4_K_M

llama.cpp

llama-server -hf Qwen/Qwen3-32B-GGUF:Q4_K_M

Then open http://localhost:8080 for a chat window.

MLX (Mac)

pip install -U mlx-lm
mlx_lm.chat --model mlx-community/Qwen3-32B-4bit

vLLM (GPU server)

vllm serve Qwen/Qwen3-32B

Full-precision weights, for a server GPU.

Open in LM Studio

Memory needed: weights, context cache and runtime buffers.

Quantization	4K context	32K context	131K context
16-bit F16	68 GB	75 GB	101 GB
8-bit Q8_0	37 GB	45 GB	70 GB
4-bit Q4_K_M	22 GB	30 GB	55 GB

Check it against your GPU, Mac or phone →