# Claude Sonnet 5.5 vs Kimi K3

As of October 21, 2026, Kimi K3 costs 50% more per input token than Claude Sonnet 5.5 ($3.00 vs $2.00 per 1M) and has a 1,048,576-token context window versus 1,000,000.

- Page: https://models.litellm.ai/compare/claude-sonnet-5-5-vs-kimi-k3
- Claude Sonnet 5.5: https://models.litellm.ai/models/claude-sonnet-5-5
- Kimi K3: https://models.litellm.ai/models/kimi-k3

## Side by side

| | Claude Sonnet 5.5 | Kimi K3 |
|---|---|---|
| Input, per 1M tokens | $2.00 | $3.00 |
| Output, per 1M tokens | $10.00 | $15.00 |
| Cache read, per 1M tokens | $0.10 | $0.30 |
| Cache write, per 1M tokens | $2.50 | – |
| Context window | 1,000,000 tokens | 1,048,576 tokens |
| Max output | 128,000 tokens | 1,048,576 tokens |
| Image input | Yes | Yes |
| Tool calling | Yes | Yes |
| Structured output (JSON) | Yes | Yes |
| Prompt caching | Yes | Yes |
| Reasoning | Yes | Yes |
| Providers on LiteLLM | 7 | 15 |
| Price shown is from | Anthropic | Moonshot AI |
| Added to LiteLLM | September 28, 2026 | August 16, 2026 |

Headline prices in USD per 1M tokens, from LiteLLM's model price file.

## Cheapest provider for each

- Claude Sonnet 5.5: Anthropic at $2.00 input per 1M tokens (`claude-sonnet-5-5`)
- Kimi K3: Sail at $2.50 input per 1M tokens (`sail/moonshotai/Kimi-K3`)

## Benchmark scores

| Benchmark | Claude Sonnet 5.5 | Kimi K3 |
|---|---|---|
| Epoch Capabilities Index (Overall) | 166.2 | 157.9 |
| FrontierCode (Coding and agents) | 52.1% (xhigh) | 44.2% |
| FrontierSWE (Coding and agents) | 61.9% (max) | 25.9% (max) |
| APEX-Agents (Coding and agents) | 75.5% (max) | 47.8% |
| GPQA Diamond (Reasoning and knowledge) | 95.6% (max) | 93.1% (max) |
| SimpleQA Verified (Reasoning and knowledge) | 46.5% (max) | 50.6% (max) |
| FrontierMath (tiers 1-3) (Math) | 88.8% (max) | 72.2% (max) |
| FrontierMath (tier 4) (Math) | 80.5% (max) | 39.0% (max) |
| OTIS Mock AIME 2024-2025 (Math) | 100.0% (max) | 97.2% (max) |

From the Epoch AI Benchmarking Hub (https://epoch.ai/benchmarks), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/), data as of September 29, 2026: Epoch's own runs and results it collects from each benchmark's leaderboard. Higher is better.

## When to pick which

- Claude Sonnet 5.5: Costs 33% less per input token: $2.00 vs $3.00 per 1M.
- Kimi K3: Returns up to 1,048,576 tokens per response, vs 128,000.
- Kimi K3: Available from 15 providers through LiteLLM, vs 7.

## FAQ

### Which is cheaper, Claude Sonnet 5.5 or Kimi K3?

Claude Sonnet 5.5 is cheaper at 3 input tokens per output token: $4.00 per 1M tokens blended ($2.00 input, $10.00 output), versus $6.00 ($3.00 and $15.00) for Kimi K3, as of October 21, 2026.

### Which has the larger context window, Claude Sonnet 5.5 or Kimi K3?

Kimi K3 has the larger context window: 1,048,576 tokens, versus 1,000,000 for Claude Sonnet 5.5.

### Can I switch from Claude Sonnet 5.5 to Kimi K3 without changing my code?

Yes. Through LiteLLM both use the OpenAI format, so switching means changing the model name from "claude-sonnet-5-5" to "moonshot/kimi-k3", or pointing a LiteLLM gateway model alias at the other model. You can also use one as a fallback for the other.