Claude Haiku 4.5 vs Kimi K2.7 Code
Side-by-side API pricing and Artificial Analysis performance for Claude Haiku 4.5 (Anthropic) and Kimi K2.7 Code (Moonshot AI). Green cells mark the better value in each row.
Prices updated Jul 22, 2026, 1:39 AM UTC · refreshed hourly
| Claude Haiku 4.5 Anthropic | Kimi K2.7 Code Moonshot AI | |
|---|---|---|
| Pricing | ||
| Input $/1M | $1.00 | $0.82 |
| Output $/1M | $5.00 | $3.75 |
| Blended $/1M (3:1) | $2.00 | $1.55 |
| RAG example (30K in / 2K out) | $0.04 | $0.0321 |
| Quality & speed | ||
| Intelligence Index | 29.6 | 41.9 |
| Coding Index | 43.9 | 60.8 |
| Agentic Index | 16.4 | 29.6 |
| Output speed (tok/s) | — | — |
| Time to first token | — | — |
| Specs | ||
| Context window | 200K | 262K |
| Input modalities | text, image, file | text, image |
FAQ
Which is cheaper, Claude Haiku 4.5 or Kimi K2.7 Code?
Kimi K2.7 Code has the lower blended API price at $1.55 per 1M tokens (3:1 input:output mix), versus $2.00 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), Kimi K2.7 Code costs about $0.0321 per request versus $0.04. Also compare context windows: Claude Haiku 4.5 offers 200K and Kimi K2.7 Code offers 262K.
Which scores higher on Artificial Analysis benchmarks?
Kimi K2.7 Code leads on the Artificial Analysis Intelligence Index (41.9 vs 29.6). Check coding and agentic indices on this page for workload-specific tradeoffs.