Claude Sonnet 5 vs gpt-oss-120b
Side-by-side API pricing and Artificial Analysis performance for Claude Sonnet 5 (Anthropic) and gpt-oss-120b (OpenAI). Green cells mark the better value in each row.
Prices updated Sep 19, 2026, 7:46 PM UTC · refreshed hourly
| Claude Sonnet 5 Anthropic | gpt-oss-120b OpenAI | |
|---|---|---|
| Pricing | ||
| Input $/1M | $2.00 | $0.15 |
| Output $/1M | $10.00 | $0.60 |
| Blended $/1M (3:1) | $4.00 | $0.262 |
| RAG example (30K in / 2K out) | $0.08 | $0.0057 |
| Quality & speed | ||
| Intelligence Index | 38.4 | 12.3 |
| Coding Index | 71.5 | 30.4 |
| Agentic Index | 44.3 | 6.2 |
| Output speed (tok/s) | 86 | 195 |
| Time to first token | 120.21s | 0.53s |
| Specs | ||
| Context window | 1M | 131K |
| Input modalities | text, image, file | text |
FAQ
Which is cheaper, Claude Sonnet 5 or gpt-oss-120b?
gpt-oss-120b has the lower blended API price at $0.262 per 1M tokens (3:1 input:output mix), versus $4.00 for the other model. Prices are live from OpenRouter and refresh about hourly.
Which is better for RAG workloads?
For a typical RAG request (30K input / 2K output tokens), gpt-oss-120b costs about $0.0057 per request versus $0.08. Also compare context windows: Claude Sonnet 5 offers 1M and gpt-oss-120b offers 131K.
Which scores higher on Artificial Analysis benchmarks?
Claude Sonnet 5 leads on the Artificial Analysis Intelligence Index (38.4 vs 12.3). Check coding and agentic indices on this page for workload-specific tradeoffs.