Qwen2.5 72B Instruct
by Alibaba (Qwen) · qwen/qwen-2.5-72b-instruct · #96 cheapest of 325 paid models
Prices updated Jul 22, 2026, 1:33 AM UTC · refreshed hourly
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...
Compare vs:Claude Sonnet 5Claude Opus 4.6Claude Haiku 4.5GPT-5.6 Sol
Pricing
Input
$0.36
per 1M tokens
Output
$0.40
per 1M tokens
Blended (3:1)
$0.37
3 input : 1 output
Cache read
—
per 1M cached tokens
Cache write
—
per 1M tokens
Cache & batch economics
Effective prices for Qwen2.5 72B Instruct (no cache-read rate published — hits billed as normal input).
| Cache hit rate | Effective blended $/1M | Example request* | vs no cache |
|---|---|---|---|
| 0% | $0.37 | $0.0371 | — |
| 50% | $0.37 | $0.0371 | −0% |
| 90% | $0.37 | $0.0371 | −0% |
*Example: 100K sticky context + 2K new input + 1K output. Batch discount is an approximate 50% off list rates (available for OpenAI / Anthropic / Google — not flagged for this creator). Confirm on the vendor’s pricing page.
What a request costs
| Workload | Input tokens | Output tokens | Cost / request | Cost / 1K requests |
|---|---|---|---|---|
| Short chat message | 500 | 300 | $0.0003 | $0.3 |
| Document summary | 8,000 | 1,000 | $0.00328 | $3.28 |
| Codebase question (RAG) | 30,000 | 2,000 | $0.0116 | $11.60 |
| Long-context analysis | 150,000 | 5,000 | $0.056 | $56.00 |
Performance
Intelligence Index
9.6
AA composite quality
Coding Index
—
Math Index
14
Agentic Index
—
Output speed
0 tok/s
median
Time to first token
0s
median
Time to first answer
0s
after reasoning tokens
Artificial Analysis benchmarks
| Benchmark | Score |
|---|---|
| MMLU-Pro | 72.0% |
| GPQA Diamond | 49.1% |
| Humanity's Last Exam | 4.2% |
| LiveCodeBench | 27.6% |
| SciCode | 26.7% |
| MATH-500 | 85.8% |
| AIME | 16.0% |
| AIME 2025 | 14.0% |
| IFBench | 36.9% |
| AA-LCR (Long Context Reasoning) | 20.3% |
| Terminal-Bench Hard | 4.5% |
| τ²-Bench (Telecom) | 34.5% |
Specs
Context window
33K
tokens
Max output
16K
tokens
Input modalities
text
Tokenizer
Qwen