Claude Haiku 4.5 vs DeepSeek V4 Pro 0423
Side-by-side API pricing and Artificial Analysis performance for Claude Haiku 4.5 (Anthropic) and DeepSeek V4 Pro 0423 (DeepSeek). Green cells mark the better value in each row.
Prices updated Sep 16, 2026, 3:01 AM UTC · refreshed hourly
| Claude Haiku 4.5 Anthropic | DeepSeek V4 Pro 0423 DeepSeek | |
|---|---|---|
| Pricing | ||
| Input $/1M | $1.00 | $1.60 |
| Output $/1M | $5.00 | $3.20 |
| Blended $/1M (3:1) | $2.00 | $2.00 |
| RAG example (30K in / 2K out) | $0.04 | $0.0544 |
| Quality & speed | ||
| Intelligence Index | 17.6 | 36.3 |
| Coding Index | 43.9 | 68.8 |
| Agentic Index | 10.3 | 27.7 |
| Output speed (tok/s) | — | 87 |
| Time to first token | — | 1.13s |
| Specs | ||
| Context window | 200K | 1.05M |
| Input modalities | text, image, file | text |
FAQ
Which is cheaper, Claude Haiku 4.5 or DeepSeek V4 Pro 0423?
Claude Haiku 4.5 and DeepSeek V4 Pro 0423 currently have the same blended API price of $2.00 per 1M tokens.
Which is better for RAG workloads?
For a typical RAG request (30K input / 2K output tokens), Claude Haiku 4.5 costs about $0.04 per request versus $0.0544. Also compare context windows: Claude Haiku 4.5 offers 200K and DeepSeek V4 Pro 0423 offers 1.05M.
Which scores higher on Artificial Analysis benchmarks?
DeepSeek V4 Pro 0423 leads on the Artificial Analysis Intelligence Index (36.3 vs 17.6). Check coding and agentic indices on this page for workload-specific tradeoffs.