Skip to content
Weights Not Walls
Menu

moonshotai

Kimi-K2.7-Code

Reviewedaccepts images
Size
1.0T parametersfrom the GGUF metadata of a conversion, because this repo's own weights are packed: unsloth/Kimi-K2.7-Code-GGUF
Licence
unknown, MiniMax Model LicenseOur reading, check the licence: Despite the id, this text is the MiniMax Model License (dated for MiniMax-M2.5), not a modified MIT licence: it lets you use, copy, change and distribute the model with no limit on commercial use and no user or revenue threshold. If you distribute the model or a product or service that uses it, you must include the licence, mark changed files and keep a Notice file, though you may license your own changes differently. A Prohibited Uses Policy is part of the licence and bars, among other things, any military purpose and fully automated decisions that adversely affect someone's legal rights.
On the Hub since
11 Jun 2026the date the repo was created
Context length
256k tokensfrom the model's config
Kind
Base model (no base model declared on the Hub)KimiK25ForConditionalGeneration
On Hugging Face
moonshotai/Kimi-K2.7-Code323k downloads in the last 30 days, 1.4k likes

Scores

14 independent results on 14 eval sets
Chess PuzzlesRe-run by Epoch AI
  • default setting14.0%±3.513 Oct 2026, log

Epoch's name: kimi-k2.7-code. Best closed model on this set: claude-fable-5-2_max, 79.0% (Anthropic). All 166 results

  • default setting12.2%±5.213 Jun 2026, log

Epoch's name: kimi-k2.7-code. Best closed model on this set: claude-fable-5-2_max, 100.0% (Anthropic). All 83 results

  • default setting54.0%±3.013 Jun 2026, log

Epoch's name: kimi-k2.7-code. Best closed model on this set: claude-fable-5-2_max, 95.1% (Anthropic). All 127 results

GPQA diamondRe-run by Epoch AI
  • default setting87.9%±2.37 Aug 2026, log

Epoch's name: kimi-k2.7-code. Best closed model on this set: gpt-6.2-astra_max, 96.8% (OpenAI). All 341 results

OTIS Mock AIME 2024-2025Re-run by Epoch AI
  • default setting95.6%±3.17 Aug 2026, log

Epoch's name: kimi-k2.7-code. Best closed model on this set: claude-fable-5_high, 100.0% (Anthropic). All 318 results

SimpleQA VerifiedRe-run by Epoch AI
  • default setting36.5%±1.527 Aug 2026, log

Epoch's name: kimi-k2.7-code. Best closed model on this set: claude-fable-5-2_max, 94.0% (Anthropic). All 105 results

APEX-AgentsReported

Epoch's name: kimi-k2.7-code. Best closed model on this set: gemini-4-carbon_unknown, 84.2% (Google DeepMind). All 66 results

CritPtReported

Epoch's name: kimi-k2.7-code. Best closed model on this set: gemini-4-carbon_high, 36.0% (Google DeepMind). All 271 results

DeepSWEReported

Epoch's name: kimi-k2.7-code. Best closed model on this set: gpt-6-astra_xhigh, 74.1% (OpenAI). All 69 results

FrontierCodeReported
  • Harness: mini-swe-agent; Reasoning effort: none30.1%cognition.com

Epoch's name: kimi-k2.7-code. Best closed model on this set: claude-opus-5-5_medium, 54.6% (Anthropic). All 46 results

GBAEvalReported

Epoch's name: kimi-k2.7-code. Best closed model on this set: claude-opus-5-5_unknown, 90.6% (Anthropic). All 28 results

SimpleBenchReported

Epoch's name: kimi-k2.7-code. Best closed model on this set: claude-fable-5-2_unknown, 94.9% (Anthropic). All 125 results

Epoch's name: kimi-k2.7-code. Best closed model on this set: claude-fable-5_high, 95.0% (Anthropic). All 29 results

WeirdMLReported

Epoch's name: kimi-k2.7-code. Best closed model on this set: gpt-6-astra_promax, 93.6% (OpenAI). All 173 results

Claimed by the maker Claimed

The model card has no claimed scores in a machine-readable form (the card's model-index), so none are shown. The maker may give scores in the card text: read the card.

Other views, which we link to and do not copy: look this model up on Artificial Analysis and the Arena leaderboard.

Independent scores from Epoch AI (CC-BY 4.0), fetched 19 Nov 2026: Epoch AI, ‘Capabilities & benchmarking’. Published online at epoch.ai. Retrieved from ‘https://epoch.ai/benchmarks’ [online resource]. Each row links its own source. What the badges mean.

Run it yourself

file sizes read from the Hub 19 Nov 2026

Memory needed = weights + context cache + 0.50 GB for the runtime. Context cache at 8k tokens: 0.54 GB, worked out from the model's config (moonshotai/Kimi-K2.7-Code), at 16-bit precision. Choose a machine above to see a verdict for each format. Method

GGUF for llama.cpp, Ollama, LM Studio and others

unsloth/Kimi-K2.7-Code-GGUF (converted by unsloth), 21k downloads. Sizes are the real files on the Hub, all parts added up. For image input you also need the projector file mmproj-F16.gguf (0.89 GB), not counted below.

FormatSizeMemory neededGet it
UD-IQ1_M283.0 GBfile size, 8 parts284.1 GB283.0 + 0.54 + 0.50
Commands
llama.cpp
llama-server -hf unsloth/Kimi-K2.7-Code-GGUF:UD-IQ1_M
Download
hf download unsloth/Kimi-K2.7-Code-GGUF --include "UD-IQ1_M/Kimi-K2.7-Code-UD-IQ1_M-*-of-00008.gguf" --local-dir Kimi-K2.7-Code-GGUF
UD-IQ2_XXS296.0 GBfile size, 8 parts297.0 GB296.0 + 0.54 + 0.50
Commands
llama.cpp
llama-server -hf unsloth/Kimi-K2.7-Code-GGUF:UD-IQ2_XXS
Download
hf download unsloth/Kimi-K2.7-Code-GGUF --include "UD-IQ2_XXS/Kimi-K2.7-Code-UD-IQ2_XXS-*-of-00008.gguf" --local-dir Kimi-K2.7-Code-GGUF
UD-IQ2_M296.1 GBfile size, 8 parts297.2 GB296.1 + 0.54 + 0.50
Commands
llama.cpp
llama-server -hf unsloth/Kimi-K2.7-Code-GGUF:UD-IQ2_M
Download
hf download unsloth/Kimi-K2.7-Code-GGUF --include "UD-IQ2_M/Kimi-K2.7-Code-UD-IQ2_M-*-of-00008.gguf" --local-dir Kimi-K2.7-Code-GGUF
UD-Q2_K_XL316.2 GBfile size, 8 parts317.2 GB316.2 + 0.54 + 0.50
Commands
llama.cpp
llama-server -hf unsloth/Kimi-K2.7-Code-GGUF:UD-Q2_K_XL
Download
hf download unsloth/Kimi-K2.7-Code-GGUF --include "UD-Q2_K_XL/Kimi-K2.7-Code-UD-Q2_K_XL-*-of-00008.gguf" --local-dir Kimi-K2.7-Code-GGUF
UD-IQ3_S390.0 GBfile size, 10 parts391.1 GB390.0 + 0.54 + 0.50
Commands
llama.cpp
llama-server -hf unsloth/Kimi-K2.7-Code-GGUF:UD-IQ3_S
Download
hf download unsloth/Kimi-K2.7-Code-GGUF --include "UD-IQ3_S/Kimi-K2.7-Code-UD-IQ3_S-*-of-00010.gguf" --local-dir Kimi-K2.7-Code-GGUF
UD-Q3_K_M431.8 GBfile size, 11 parts432.8 GB431.8 + 0.54 + 0.50
Commands
llama.cpp
llama-server -hf unsloth/Kimi-K2.7-Code-GGUF:UD-Q3_K_M
Download
hf download unsloth/Kimi-K2.7-Code-GGUF --include "UD-Q3_K_M/Kimi-K2.7-Code-UD-Q3_K_M-*-of-00011.gguf" --local-dir Kimi-K2.7-Code-GGUF
UD-Q3_K_XL432.0 GBfile size, 11 parts433.1 GB432.0 + 0.54 + 0.50
Commands
llama.cpp
llama-server -hf unsloth/Kimi-K2.7-Code-GGUF:UD-Q3_K_XL
Download
hf download unsloth/Kimi-K2.7-Code-GGUF --include "UD-Q3_K_XL/Kimi-K2.7-Code-UD-Q3_K_XL-*-of-00011.gguf" --local-dir Kimi-K2.7-Code-GGUF
UD-IQ4_XS461.1 GBfile size, 12 parts462.1 GB461.1 + 0.54 + 0.50
Commands
llama.cpp
llama-server -hf unsloth/Kimi-K2.7-Code-GGUF:UD-IQ4_XS
Download
hf download unsloth/Kimi-K2.7-Code-GGUF --include "UD-IQ4_XS/Kimi-K2.7-Code-UD-IQ4_XS-*-of-00012.gguf" --local-dir Kimi-K2.7-Code-GGUF
UD-Q4_K_XL543.6 GBfile size, 14 parts544.7 GB543.6 + 0.54 + 0.50
Commands
llama.cpp
llama-server -hf unsloth/Kimi-K2.7-Code-GGUF:UD-Q4_K_XL
Download
hf download unsloth/Kimi-K2.7-Code-GGUF --include "UD-Q4_K_XL/Kimi-K2.7-Code-UD-Q4_K_XL-*-of-00014.gguf" --local-dir Kimi-K2.7-Code-GGUF
UD-Q8_K_XL553.7 GBfile size, 14 parts554.7 GB553.7 + 0.54 + 0.50
Commands
llama.cpp
llama-server -hf unsloth/Kimi-K2.7-Code-GGUF:UD-Q8_K_XL
Download
hf download unsloth/Kimi-K2.7-Code-GGUF --include "UD-Q8_K_XL/Kimi-K2.7-Code-UD-Q8_K_XL-*-of-00014.gguf" --local-dir Kimi-K2.7-Code-GGUF

MLX for Apple silicon

FormatSizeMemory neededRepo
MLX 6bitspicyneuron/Kimi-K2.7-Code-MLX-3.6bit427.6 GBfile size428.6 GB427.6 + 0.54 + 0.50Download page
MLXmlx-community/Kimi-K2.7-Code-mlx-DQ3_K_M-q8438.3 GBfile size439.3 GB438.3 + 0.54 + 0.50Download page

Original weights

The maker's own weights are packed into integer tensors, so the safetensors header gives no size we can trust; the repo lists the real file sizes.

hf download moonshotai/Kimi-K2.7-Code --local-dir Kimi-K2.7-Code
Other conversions (5; sizes not read)

The Hub tags 29 quantised repos as conversions of this model; we read the most downloaded and show the maker's own first.

Run it by API

Hugging Face Inference Providers, fetched 19 Nov 2026, 13:58 UTC

9 providers serve this model through the Hugging Face router. Prices are US dollars per million tokens. Precision is what the provider reports; a lower precision than the original weights can change answers. Live first, then cheapest output.

ProviderInOutContextPrecisionTokens/sFirst tokenUptime 24hToolsJSON
CoreWeave$0.60$3.00256kint41080.8 s99.3%yesyes
deepinfra$0.68$3.40256knot given430.8 s?yesyes
Inceptron$0.66$3.50256kint4650.8 s100.0%yesyes
Venice$0.75$3.50250kint4432.9 s100.0%yesyes
SiliconFlow$0.86$3.80256kfp8124.1 s100.0%yesno
Novita$0.91$3.84256kint4322.1 s100.0%yesyes
Cloudflare$0.95$4.00256knot given311.3 s100.0%yesyes
Moonshot AIthe maker$0.95$4.00256knot given271.8 s99.8%yesyes
featherless-ai???not given???not givennot given

Through the router, with a Hugging Face token (the router bills your Hugging Face account). Change the name after the colon to pick another provider.

curl https://router.huggingface.co/v1/chat/completions \ -H "Authorization: Bearer $HF_TOKEN" -H "Content-Type: application/json" \ -d '{"model": "moonshotai/Kimi-K2.7-Code:coreweave", "messages": [{"role": "user", "content": "Hello"}]}'

Not built yet: one key for every open model, with BlockRun, and our own tested-quality score per provider. Until then, precision and measured speed are the best guide to a provider here.

Fine-tunes of this model

from our snapshot of the Hub

Repos that declare this model as their base, most downloaded first. Listing is not a recommendation; unreviewed ones have not been checked.

Planned for this page Not built yet

Try it in private chat
A link to Solana AI's confidential chat for the models it serves. This model is not listed as served there.
Fine-tune it
Fine-tuning a few base models from this page and publishing the result here.
Creators and earnings
Pages for people who publish fine-tunes and adapters of this model, and what they earn.
Fund and bounties
Bounties for new open models, paid from a public fund. How the fund will work.
Contribute compute
How to give compute to open model work, with Akash.

Model data from the Hugging Face Hub 19 Nov 2026, 13:43 UTC. The repo. Found a mistake on this page? Tell us.