Baltor Get started

Models

Kimi-K2.7-Code

By moonshotai. 1.03 trillion parameters, a context window of 262,144 tokens and the licence other.

Facts

Released
2026-06-11 the day the repository was first published on Hugging Face Hugging Face, read
Licence
other the licence the model card declares Hugging Face, read
Open weights
Yes the weights are published in this Hugging Face repository Hugging Face, read
Parameters
1.03 trillion counted from the safetensors weight files Hugging Face, read
Active parameters
33.3 billion estimate estimated from the expert counts and sizes in the configuration model configuration, read
Knowledge cutoff
Unknown
Context window
  • 262,144 tokens the maximum position embeddings in the model configuration model configuration, read
Longest output
Unknown
Tool calling
Unknown
Structured output
Unknown
Reasoning controls
Unknown
Inputs and outputs
text, image in, text out Hugging Face, read
Good for
  • Vision: Hugging Face files it under image-text-to-text Hugging Face, read
  • Coding: the maker's name for the model marks it for code Hugging Face, read

Prices

Each provider's own price in US dollars per million tokens, as an openly licensed source lists it
ProviderInputOutputContextAs ofSource
Inceptron0.660 USD3.40 USD262,144 older than 30 days; check the providermodels.dev, read
Kilo Gateway0.671 USD3.35 USD262,144 older than 30 days; check the providermodels.dev, read
NanoGPT0.671 USD3.35 USD262,144 older than 30 days; check the providermodels.dev, read
OpenRouter0.671 USD3.35 USD262,144 older than 30 days; check the providermodels.dev, read
Deep Infra0.680 USD3.40 USD262,144 older than 30 days; check the providermodels.dev, read
IteraCompute0.690 USD3.29 USD262,144 older than 30 days; check the providermodels.dev, read
Ambient0.690 USD3.49 USD262,144 older than 30 days; check the providermodels.dev, read
CoreWeave0.710 USD3.50 USD262,144 older than 30 days; check the providermodels.dev, read
ai&0.750 USD3.50 USD262,144 older than 30 days; check the providermodels.dev, read
SiliconFlow0.859 USD3.80 USD262,144 older than 30 days; check the providermodels.dev, read
inferenceapis0.890 USD4.42 USD262,144LiteLLM, read
Abacus0.950 USD4.00 USD262,144 older than 30 days; check the providermodels.dev, read
Baseten0.950 USD4.00 USD262,000 older than 30 days; check the providermodels.dev, read
CommandCode0.950 USD4.00 USD256,000 older than 30 days; check the providermodels.dev, read
HPC-AI0.950 USD4.00 USD256,000 older than 30 days; check the providermodels.dev, read
Hugging Face0.950 USD4.00 USD262,144 older than 30 days; check the providermodels.dev, read
Nebius Token Factory0.950 USD4.00 USD262,144 older than 30 days; check the providermodels.dev, read
Novita AI0.950 USD4.00 USD262,144 older than 30 days; check the providermodels.dev, read
Ofox0.950 USD4.00 USD262,144 older than 30 days; check the providermodels.dev, read
Pioneer0.950 USD4.00 USD256,000 older than 30 days; check the providermodels.dev, read
Together AI0.950 USD4.00 USD262,144LiteLLM, read
Vercel AI Gateway0.950 USD4.00 USD256,000 older than 30 days; check the providermodels.dev, read
ZenMux0.950 USD4.00 USD262,144 older than 30 days; check the providermodels.dev, read
OpenReason1.00 USD4.22 USD262,144 older than 30 days; check the providermodels.dev, read
TensorX1.25 USD4.50 USD262,144 older than 30 days; check the providermodels.dev, read

Run it on your own hardware

Weights are the sizes of the files a source lists, or an estimate from the parameter count where none does. Memory is an estimate: weights plus KV cache plus 512 MiB and 5 percent of the weights for runtime buffers. Check it against your hardware.

QuantizationWeightsMemory at 8,192 tokensMemory at 32,768 tokens
UD-IQ1_M283 GiB298 GiB300 GiB
UD-IQ2_XXS296 GiB312 GiB313 GiB
UD-IQ2_M296 GiB312 GiB314 GiB
UD-Q2_K_XL316 GiB333 GiB335 GiB
UD-IQ3_S390 GiB411 GiB412 GiB
UD-Q3_K_M432 GiB454 GiB456 GiB
UD-Q3_K_XL432 GiB455 GiB456 GiB
UD-Q4_K_XL544 GiB572 GiB573 GiB

Start it with a local runtime

  • Ollama
    ollama run hf.co/unsloth/Kimi-K2.7-Code-GGUF:UD-IQ1_M
  • LM Studio
    lms get unsloth/Kimi-K2.7-Code-GGUF
    lms server start
  • llama.cpp server
    llama-server -hf unsloth/Kimi-K2.7-Code-GGUF:UD-IQ1_M --jinja
  • vLLM
    vllm serve moonshotai/Kimi-K2.7-Code
  • SGLang
    sglang serve --model-path moonshotai/Kimi-K2.7-Code --port 30000

Use it from your harness

Set up for Abacus with the model moonshotai/Kimi-K2.7-Code. Each endpoint page has the same setup for its own address.

OpenCode

Put this in opencode.json in your project folder:

{
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "abacus": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "Abacus",
      "options": {
        "baseURL": "https://routellm.abacus.ai/v1",
        "apiKey": "{env:ABACUS_API_KEY}"
      },
      "models": {
        "moonshotai/Kimi-K2.7-Code": {
          "name": "Kimi-K2.7-Code"
        }
      }
    }
  }
}
  • OpenCode reads any OpenAI-compatible address through the @ai-sdk/openai-compatible package, and an address that speaks the Responses API through @ai-sdk/openai.

From OpenCode documentation, read .

Pi

Put this in ~/.pi/agent/models.json:

{
  "providers": {
    "abacus": {
      "baseUrl": "https://routellm.abacus.ai/v1",
      "api": "openai-completions",
      "apiKey": "$ABACUS_API_KEY",
      "models": [
        {
          "id": "moonshotai/Kimi-K2.7-Code"
        }
      ]
    }
  }
}
  • The apiKey field can name an environment variable as $NAME.

From Pi documentation, read .

Codex

Codex speaks only the Responses API, and Abacus documents no Responses address. A gateway that offers one can sit in between.

  • Codex speaks the Responses API only: responses is the one supported wire API of a custom provider. Ollama and LM Studio are built in and start with --oss.

From Codex documentation, read .

Claude Code

Claude Code sends Anthropic Messages requests, and Abacus documents no such address. A gateway that translates to that API can sit in between.

  • Claude Code sends Anthropic Messages requests to ANTHROPIC_BASE_URL. Anthropic says it does not support routing Claude Code to models other than Claude through any gateway, so some features may not work with another model.

From Claude Code documentation, read .

Other ways to reach it

Published results

These are results other people published. Baltor did not run them and does not rank models by them.

Two more publishers are linked, never copied, because their terms do not allow it:

  • Artificial Analysis publishes its own intelligence, speed and price measurements.
  • OpenRouter lists current prices and uptime for each provider it routes to.