Claude Haiku 4.5 vs Kimi K2.6: API pricing
Anthropic's Claude Haiku 4.5 against Moonshot AI's Kimi K2.6 on price alone — per-token rates, caching, and context window. Run your own workload through both with the calculator below; capability differences are yours to judge on your evals.
| Claude Haiku 4.5 | Kimi K2.6 | |
|---|---|---|
| Input / 1M tokens | $1.00 | $0.95 |
| Cached input / 1M | $0.10 | $0.16 |
| Output / 1M tokens | $5.00 | $4.00 |
| Context window | 200K | 256K |
| Max output | 64K | 256K |
| Vendor | Anthropic | Moonshot AI |
Prices last synced Aug 20, 2026 from the LiteLLM community price catalog — the same catalog Marginal prices real usage events against, refreshed daily. Spot an error? Tell us.
Price takeaways
- Input: effectively the same price.
- Output: Kimi K2.6 is 1.3x cheaper.
- Cached input: Claude Haiku 4.5 is 1.6x cheaper.
- Ratios shift with workload shape — output-heavy traffic weights the output rate, cache-friendly traffic weights the cached rate. The calculator below does the arithmetic on your numbers.
Your workload on both
Your monthly workload
Cache hit rate = share of input tokens served from the provider's prompt cache, billed at the model's cached-input price. Estimates use standard per-token rates — batch discounts, service tiers, and long-context surcharges are not applied.
| Model | Input | Cached input | Output | Per request | Monthly |
|---|---|---|---|---|---|
| Kimi K2.6cheapest | $570 | $64.00 | $1,000 | $0.00327 | $1,634 |
| Claude Haiku 4.5 | $600 | $40.00 | $1,250 | $0.00378 | $1,890 |
Related comparisons
Switching models? Watch the bill move in real time.
Marginal prices every call against that day's rates, so an A/B between these two shows up as dollars per customer and feature — not a surprise at month end.