Claude Fable 5$22.000/MClaude Opus 4.8$11.000/MClaude Opus 4.7$11.000/MClaude Opus 4.6$11.000/MClaude Opus 4.5$33.000/MClaude Sonnet 3.7$6.600/MClaude Opus 3$33.000/MClaude 2.1$12.800/MClaude 2$12.800/MGPT-5.5$12.500/MGPT-5.2$5.425/MGPT-5.2-Codex$5.425/MGPT-5$3.875/MGPT-4.5$97.500/MGPT-4 Turbo Preview$16.000/MGPT-4$39.000/MGPT-4-32k$78.000/Mo3$19.000/Mo3-mini$2.090/Mo4-mini$2.090/Mo1$28.500/Mo1-mini$5.700/Mo1-preview$28.500/MGemini 3.5 Pro$5.000/MGemini 3.1 Pro$5.000/MGemini 3 Pro$5.000/MGemini 2.5 Pro$3.875/MGemini 1.5 Pro$2.375/MGemini 1.0 Ultra$12.000/MGemini 1.0 Pro$0.800/MClaude Fable 5$22.000/MClaude Opus 4.8$11.000/MClaude Opus 4.7$11.000/MClaude Opus 4.6$11.000/MClaude Opus 4.5$33.000/MClaude Sonnet 3.7$6.600/MClaude Opus 3$33.000/MClaude 2.1$12.800/MClaude 2$12.800/MGPT-5.5$12.500/MGPT-5.2$5.425/MGPT-5.2-Codex$5.425/MGPT-5$3.875/MGPT-4.5$97.500/MGPT-4 Turbo Preview$16.000/MGPT-4$39.000/MGPT-4-32k$78.000/Mo3$19.000/Mo3-mini$2.090/Mo4-mini$2.090/Mo1$28.500/Mo1-mini$5.700/Mo1-preview$28.500/MGemini 3.5 Pro$5.000/MGemini 3.1 Pro$5.000/MGemini 3 Pro$5.000/MGemini 2.5 Pro$3.875/MGemini 1.5 Pro$2.375/MGemini 1.0 Ultra$12.000/MGemini 1.0 Pro$0.800/M
BETA
Subscribe
← Back to Index·All Qwen models
Qwen
QwenEfficient

Qwen2.5-32B-Instruct

Balanced

Mid-size Qwen2.5 — single-GPU friendly while keeping most of the quality of the 72B flagship.

Qwen2.5-32B-Instruct is a efficient AI model from Qwen. It costs $0.700 per million input tokens and $0.700 per million output tokens (blended $0.700/M), with a 128K-token context window.

Released Sep 2024Modalities textOfficial model page ↗Compare with another model →Estimate monthly cost →
INPUT
$0.700/M
per million input tokens
OUTPUT
$0.700/M
per million output tokens
BLENDED 70/30
$0.700/M
default reference rate · how it's calculated →
CONTEXT
128K
128,000 tokens
What it's good at
  • Open weights
  • Single-GPU friendly
  • Strong multilingual
  • 128K context
Typical use cases
  • On-prem multilingual chat
  • Fine-tune base
  • Cost-sensitive RAG
Benchmarks
vs. best public score
MMLU83%
Multitask academic knowledge across 57 subjects.
GPQA Diamond49%
Graduate-level science questions, "Google-proof".
MATH83%
High-school competition math problems.
HumanEval89%
Python function synthesis from docstrings.
LMArena Elo1230 Elo
Crowd-sourced head-to-head preference Elo rating.
Hand-curated from each provider's published reports and public leaderboards. Methodology varies across sources — treat as directional rather than authoritative.
More from Qwen
See all 22 →
Qwen3.7-Max
Flagship · 1M ctx
in $2.500/Mout $7.500/M
Qwen3.7-Plus
Balanced · 1M ctx
in $0.400/Mout $1.600/M
Qwen2.5-Max
Flagship · 32K ctx
in $1.600/Mout $6.400/M
Qwen2.5-72B-Instruct
Large · 128K ctx
in $1.200/Mout $1.200/M
Qwen2.5-14B-Instruct
Small · 128K ctx
in $0.350/Mout $0.350/M
Qwen2.5-7B-Instruct
Compact · 128K ctx
in $0.100/Mout $0.100/M
Frequently asked questions

How much does Qwen2.5-32B-Instruct cost?

Qwen2.5-32B-Instruct costs $0.700 per million input tokens and $0.700 per million output tokens, for a blended reference rate of $0.700 per million tokens.

What is Qwen2.5-32B-Instruct's context window?

Qwen2.5-32B-Instruct supports up to 128K tokens of context (128,000 tokens).

What is Qwen2.5-32B-Instruct best for?

Qwen2.5-32B-Instruct is well suited to Open weights, Single-GPU friendly and Strong multilingual.

Who makes Qwen2.5-32B-Instruct?

Qwen2.5-32B-Instruct is developed and served by Qwen. It was released in Sep 2024.