Qwen2.5 72B Instruct pricing & benchmarks
Qwen2.5 72B Instruct is available from Alibaba at $0.360 per million input tokens and $0.400 per million output tokens ($0.370 blended at 3:1). It accepts up to 32,768 tokens of context. Qwen2.5 72B is the latest series of Qwen large language models. Qwen2.5 brings the following improvements upon Qwen2: - Significantly more knowledge and has greatly improved capabilities in coding and...
Pricing
Per million tokens · OpenRouter
Input$0.360
Output$0.400
Blended (3:1)$0.370
Specification
qwen/qwen-2.5-72b-instruct
Context window33K
Max output16K
ReleasedSep 2024
LicenceOpen weights
Modalitiesin:text, out:text
Benchmarks
Independently run — not vendor self-reports. Every score links its source.
| Benchmark | Score | Config | Run | Source |
|---|---|---|---|---|
| GPQA Diamond | 49.1% ±2.7 | — | 2025-01-27 | Independently run · Epoch AI |
| OTIS Mock AIME | 8.1% ±2.6 | — | 2025-02-25 | Independently run · Epoch AI |
Scores published by Epoch AI, 'AI Benchmarking Hub', used under CC BY 4.0.
Cheaper models that score at least as well on GPQA Diamond
Strictly lower blended price AND an equal-or-higher GPQA Diamond score. One benchmark is one dimension — a model that wins here may still be weaker at your specific task.
| Model | Provider | Blended $/1M | GPQA Diamond | |
|---|---|---|---|---|
| gpt-oss-120b | OpenAI | $0.070 (−81%) | 75.8% | Compare → |
| Phi 4 | Microsoft | $0.088 (−76%) | 56.1% | Compare → |
| DeepSeek V4 Flash 0731 | DeepSeek | $0.113 (−70%) | 91.0% | Compare → |
| GPT-5 Nano | OpenAI | $0.138 (−63%) | 69.4% | Compare → |
| Llama 4 Scout | Meta | $0.150 (−59%) | 51.8% | Compare → |
Where to run Qwen2.5 72B Instruct
2 hosts · all priced within ~5% of each other. Blended at 3:1. ⚠ marks an option serving less context than the best available, or re-pricing above a prompt-length threshold. Bracketed labels are the provider's own service tier or region.
| Host | Blended $/1M | In / Out | Context | Weights | Uptime 24h |
|---|---|---|---|---|---|
| DeepInfra | $0.370 | $0.360 / $0.400 | 33K | fp8 | 99.9% |
| Novita | $0.385 | $0.380 / $0.400 | 32K | bf16 | 16.0% |
Host prices and uptime from OpenRouter, synced 2026-08-08. Lower-precision weights (fp4, fp8) can cost less and score worse than the same model at bf16 — the score above is not host-specific.
Compare Qwen2.5 72B Instruct
Qwen2.5 72B Instruct vs DeepSeek V4 Flash 0731
Qwen2.5 72B Instruct vs Claude Opus 5
Qwen2.5 72B Instruct vs Gemini 3.6 Flash
Qwen2.5 72B Instruct vs Kimi K3
Qwen2.5 72B Instruct vs GPT-5.6 Luna
Qwen2.5 72B Instruct vs GPT-5.6 Terra
Qwen2.5 72B Instruct vs GPT-5.6 Sol
Qwen2.5 72B Instruct vs Grok 4.5
FAQ
How much does Qwen2.5 72B Instruct cost?
Qwen2.5 72B Instruct costs $0.360 per million input tokens and $0.400 per million output tokens on OpenRouter. At a 3:1 input:output mix that blends to $0.370 per million tokens.
What is Qwen2.5 72B Instruct's context window?
Qwen2.5 72B Instruct accepts up to 32,768 tokens of context, and can generate up to 16,384 output tokens.
How good is Qwen2.5 72B Instruct on benchmarks?
Qwen2.5 72B Instruct scores 49.1% on GPQA Diamond, independently run and published by Epoch AI. 2 benchmark results are listed on this page, each with its run date and source.
Is there a cheaper model as good as Qwen2.5 72B Instruct?
Yes — 5 models in our catalogue cost less per token than Qwen2.5 72B Instruct and score at least as high on the same benchmark. The cheapest is gpt-oss-120b at $0.070 per million tokens blended.