MoonshotAI: Kimi K2.6
by Moonshot AIMoonshotAI: Kimi K2.6 is a large language model from Moonshot AI. It costs $0.950 per million input tokens and $4.00 per million output tokens. Its context window is 262K tokens.
- Input / 1M tokens
- $0.950
- Output / 1M tokens
- $4.00
- Cached input / 1M
- $0.160
- Context window
- 262K
On repeated prefixes
tokens
Who serves it cheapest
20 hosts serve Kimi K2.6. Same weights, same API - the price difference is pure margin and routing.
| Provider | Input / 1M | Output / 1M | Context | Throughput | Uptime 24h |
|---|---|---|---|---|---|
| DigitalOceanCheapest | $0.570 | $2.40 | 262K | - | 84.3% |
| Baidufp4 | $0.580 | $2.44 | 262K | - | 100.0% |
| Decartfp4 | $0.587 | $2.47 | 262K | - | 99.9% |
| StreamLakefp8 | $0.598 | $2.52 | 256K | - | 97.4% |
| Inceptronint4 | $0.609 | $2.90 | 262K | - | 99.5% |
| Chutesint4 | $0.580 | $3.40 | 262K | - | 98.0% |
| CoreWeavefp4 | $0.650 | $3.41 | 262K | - | 99.8% |
| Crusoebf16 | $0.700 | $3.50 | 262K | - | 99.2% |
| SiliconFlowfp8 | $0.770 | $3.40 | 262K | - | 98.6% |
| DeepInfrafp4 | $0.750 | $3.50 | 262K | - | 98.8% |
| Veniceint4 | $0.750 | $3.50 | 256K | - | 96.9% |
| Parasailint4 | $0.750 | $3.50 | 262K | - | 99.2% |
| Novita | $0.800 | $3.40 | 262K | - | 98.4% |
| GMICloudfp8 | $0.855 | $3.60 | 262K | - | 94.4% |
| AtlasCloudint4 | $0.950 | $4.00 | 262K | - | 94.0% |
| Cloudflare | $0.950 | $4.00 | 262K | - | 99.8% |
| Moonshot AIint4 | $0.950 | $4.00 | 262K | - | 99.8% |
| BaseTenfp4 | $0.950 | $4.00 | 262K | - | 58.8% |
| Fireworks | $0.950 | $4.00 | 262K | - | 1.0% |
| Phala | $1.09 | $4.60 | 262K | - | 96.7% |
The spread between DigitalOcean and Phala is 1.9× for identical weights. Quantization and context limits differ, so check both columns before switching.
Benchmarks
Independent scores published alongside the catalogue.
- Coding index
- 61.8
- Agentic index
- 22.1
About Kimi K2.6
Kimi K2.6 is Moonshot AI's next-generation multimodal model, designed for long-horizon coding, coding-driven UI/UX generation, and multi-agent orchestration. It handles complex end-to-end coding tasks across Python, Rust, and Go, and...
Specifications
| Model ID | moonshotai/kimi-k2.6 |
|---|---|
| Provider | Moonshot AI |
| Context window | 262K tokens |
| Max output | 236K tokens |
| Input modalities | text, image |
| Output modalities | text |
| Knowledge cutoff | - |
| Open weights | Yes - moonshotai/Kimi-K2.6 |
| Released | April 20, 2026 |
Cheaper alternatives
Models that cost less than Kimi K2.6 while keeping at least half its context window and every input modality it supports.
Frequently asked
How much does MoonshotAI: Kimi K2.6 cost?
$0.950 per million input tokens and $4.00 per million output tokens. Cached input reads cost $0.160 per million tokens.
What is the context window of MoonshotAI: Kimi K2.6?
262K tokens, with up to 236K tokens of output per request.
Which provider serves MoonshotAI: Kimi K2.6 cheapest?
DigitalOcean at $0.570 per million input tokens - 1.9× cheaper than Phala, the most expensive of the 20 hosts serving it.
Confirm against the source: Moonshot AI official pricing.