Qwen2.5 72B Instruct
About
Instruction-optimized flagship variant for demanding production applications requiring high-accuracy complex problem-solving across industries.
Capabilities
VisionMultimodalReasoningFunction CallingTool UseStructured OutputsCode Execution
Providers(7)
Compare all →| Provider | Input (per 1M) | Output (per 1M) | Type | |
|---|---|---|---|---|
| DeepInfra | $23 | $23 | Serverless | |
| OpenRouter | $0.12 | $0.39 | Serverless | |
| Fireworks AI | $0.9 | $0.9 | Serverless | |
| Novita AI | $0.2 | $0.6 | Serverless | |
| Chutes AI | $0.18 | $0.54 | Serverless | |
| SiliconFlow | $0.28 | $0.28 | Serverless | |
| Replicate API | $1.3 | $1.3 | Serverless |
Benchmark Scores(5)
| Benchmark | Score | Version | Source |
|---|---|---|---|
| Google-Proof Q&A | 65.4 | standard | Open LLM Leaderboard |
| HellaSwag | 95.6 | standard | Open LLM Leaderboard |
| HumanEval | 92.7 | pass@1 | Open LLM Leaderboard |
| Massive Multitask Language Understanding | 88.2 | 5-shot | Open LLM Leaderboard |
| Chatbot Arena | 1270.0 | — | https://lmarena.ai |
