Claude Haiku 5.5 vs DeepSeek V4 Pro 0813: cost per task
Claude Haiku 5.5 is cheaper on 14 of 14 tasks and about 6.9x cheaper on average (geometric mean).
| Task | Claude Haiku 5.5 | DeepSeek V4 Pro 0813 | Difference |
|---|---|---|---|
| Answer a chat question | 0.07¢ | 1.3¢ | 19.2x |
| Reply to a customer email | 0.04¢ | 0.41¢ | 11.6x |
| Write a 1,500-word blog post | 0.19¢ | 0.61¢ | 3.2x |
| Write 10 product descriptions | 0.21¢ | 0.56¢ | 2.6x |
| Translate 1,000 words | 0.13¢ | 2.4¢ | 19.1x |
| Proofread one page (500 words) | 0.07¢ | 0.57¢ | 8.0x |
| Summarize a 20-page PDF | 0.22¢ | 1.3¢ | 5.7x |
| Extract data from an invoice | 0.06¢ | 0.60¢ | 9.4x |
| Analyze a 200-row spreadsheet | 0.85¢ | 3.5¢ | 4.1x |
| Answer from 5 retrieved documents (RAG) | 0.14¢ | 0.99¢ | 6.8x |
| Classify 100 support tickets | 0.09¢ | 1.1¢ | 11.5x |
| Review 500 lines of code | 0.66¢ | 3.5¢ | 5.3x |
| Write a SQL query | 0.08¢ | 0.43¢ | 5.1x |
| Run a 50-step coding agent | 6.6¢ | 30¢ | 4.6x |
Specs and list prices
| Claude Haiku 5.5 | DeepSeek V4 Pro 0813 | |
|---|---|---|
| Input per 1M | $0.10 | $0.66 |
| Output per 1M | $0.50 | $1.98 |
| Cached input per 1M | $0.01 | $0.022 |
| Context | 1M | 1M |
| Providers | 10 | 23 |
| Reasoning | default | default |
| LMArena Text | - | - |
| Released | Oct 7, 2026 | Aug 12, 2026 |
FAQ
Is Claude Haiku 5.5 or DeepSeek V4 Pro 0813 cheaper?
Claude Haiku 5.5 is cheaper on 14 of 14 tasks and about 6.9x cheaper on average (geometric mean).
How much does Claude Haiku 5.5 cost compared to DeepSeek V4 Pro 0813 per token?
Claude Haiku 5.5: $0.10 input and $0.50 output per 1M tokens. DeepSeek V4 Pro 0813: $0.66 input and $1.98 output. Token prices alone mislead - models use different numbers of tokens for the same task, which is what the task table shows.