# GPT-6.1 Terra vs Kimi K2.7 Code

As of October 27, 2026, GPT-6.1 Terra costs 5% more per input token than Kimi K2.7 Code ($1.00 vs $0.95 per 1M) and has a 922K-token context window versus 256K.

- Page: https://models.litellm.ai/compare/gpt-6.1-terra-vs-kimi-k2.7-code
- GPT-6.1 Terra: https://models.litellm.ai/models/gpt-6.1-terra
- Kimi K2.7 Code: https://models.litellm.ai/models/kimi-k2.7-code

## Side by side

| | GPT-6.1 Terra | Kimi K2.7 Code |
|---|---|---|
| Input, per 1M tokens | $1.00 | $0.95 |
| Output, per 1M tokens | $5.00 | $4.00 |
| Cache read, per 1M tokens | $0.05 | $0.19 |
| Cache write, per 1M tokens | $1.25 | – |
| Context window | 922,000 tokens | 262,144 tokens |
| Max output | 128,000 tokens | 262,144 tokens |
| Image input | Yes | Yes |
| Tool calling | Yes | Yes |
| Structured output (JSON) | Yes | Yes |
| Prompt caching | Yes | Yes |
| Reasoning | Yes | Yes |
| Providers on LiteLLM | 4 | 17 |
| Price shown is from | OpenAI | Moonshot AI |
| Added to LiteLLM | October 26, 2026 | June 2026 |

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

## Cheapest provider for each

- GPT-6.1 Terra: OpenAI at $1.00 input per 1M tokens (`gpt-6.1-terra`)
- Kimi K2.7 Code: DeepInfra at $0.68 input per 1M tokens (`deepinfra/moonshotai/Kimi-K2.7-Code`)

## When to pick which

- GPT-6.1 Terra: Larger context window: 922,000 vs 262,144 tokens.
- Kimi K2.7 Code: Costs 5% less per input token: $0.95 vs $1.00 per 1M.
- Kimi K2.7 Code: Returns up to 262,144 tokens per response, vs 128,000.

## FAQ

### Which is cheaper, GPT-6.1 Terra or Kimi K2.7 Code?

Kimi K2.7 Code is cheaper at 3 input tokens per output token: $1.712 per 1M tokens blended ($0.95 input, $4.00 output), versus $2.00 ($1.00 and $5.00) for GPT-6.1 Terra, as of October 27, 2026.

### Which has the larger context window, GPT-6.1 Terra or Kimi K2.7 Code?

GPT-6.1 Terra has the larger context window: 922,000 tokens, versus 262,144 for Kimi K2.7 Code.

### Can I switch from GPT-6.1 Terra to Kimi K2.7 Code without changing my code?

Yes. Through LiteLLM both use the OpenAI format, so switching means changing the model name from "gpt-6.1-terra" to "moonshot/kimi-k2.7-code", or pointing a LiteLLM gateway model alias at the other model. You can also use one as a fallback for the other.