Qwen3 32B

Qwen3-32B is a large language model from Alibaba's Qwen3 series. It features 32.8 billion parameters, a 128k token context window, support for 119 languages, and hybrid thinking modes allowing switching between deep reasoning and fast responses.

Arena Hard

93.8%

i
AIME 2024

81.4%

i
LiveBench

74.9%

i
MultiLF

73.0%

i
AIME 2025

72.9%

i
BFCL

70.3%

i
CodeForces

65.9%

i
LiveCodeBench

65.7%

i
Aider

50.2%

i

Pricing, uptime, and speed via OpenRouter — updated Jul 17, 2026, 04:19 AM.

Provider	Status	Input	Output	Limits	Uptime	Speed	Notes
DeepInfra	available	$0.08/Mtok	$0.28/Mtok	41K tokens context 16K tokens max output	100.0% 5m 100.0%	248 ms p50 TTFT 27 tok/s p50	fp8
Nebius	available	$0.10/Mtok	$0.30/Mtok	41K tokens context 16K tokens max output	99.7% 5m 100.0%	490 ms p50 TTFT 32 tok/s p50	fp8
Alibaba	available	$0.10/Mtok	$0.42/Mtok	131K tokens context 8K tokens max output	99.8% 5m 100.0%	356 ms p50 TTFT 85 tok/s p50
SiliconFlow	available	$0.14/Mtok	$0.57/Mtok	131K tokens context 131K tokens max output	98% 5m 100.0%	1,651 ms p50 TTFT 27 tok/s p50	fp8
Groq	available	$0.29/Mtok cache $0.14/Mtok	$0.59/Mtok	131K tokens context 41K tokens max output	100.0% 5m 100.0%	231 ms p50 TTFT 380 tok/s p50