Skip to content

Moonshot AI · Chat

Kimi K3.1 Pro

Moonshot AI 🇨🇳ChatKimi familyOpen weights. One of its sources says the weights are publicly available.The maker currently lists 4 Nov 2027 as the deprecation date. Plans can change.
From Empiriolabs's listing:
Moonshot's flagship reasoning model for long coding and agent work, with a 1M token context, image input, and web search.
llminfo IQ
142.7
Maker's price
$4.00 blended / 1M
Max context
1.1M
Listings
190

Evidence and sources

International API storefront is Singapore-based; principal operating entity 北京月之暗面科技有限公司 (Beijing).

🤗 moonshotai/Kimi-K3.1-Prollminfo IQ 142.7, ranked among chat models. From 161 benchmarks. Plenty of data behind it.How llminfo IQ is computedCoding IQ 131.3, 95% interval 121–141. From 10 benchmark families across 5 sources. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.How Coding IQ is computedFair value (49.2). Costs per task about what its benchmark results predict. From 32 benchmarks with measured task costs. Plenty of data behind it.How Value for money worksYap score: tokens used per coding task vs. models with the same resultsTypical (51 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. By effort: default 51 · max 53.Median 65,883 output tokens per task at default effort.How the Yap score worksReleased 23 Oct 2026Maker HQ: CN (source)
All rankings: #1 in Hugging Face trending today #5 on OpenRouter today #8 on OpenRouter this week #10 on OpenRouter this month #31 in Hacker News discussions this week #61 on NanoGPT recentlyAll rankings: #1 in Hugging Face trending today #5 on OpenRouter today #8 on OpenRouter this week #10 on OpenRouter this month #31 in Hacker News discussions this week #61 on NanoGPT recentlySources: Arcee AI data, Berget AI data, LiteLLM data, Models.dev data, EmpirioLabs data, llminfo researchPrimary data from Arcee AI from Berget AI: primary mode from LiteLLM: deprecation date from Models.dev: weights availability from EmpirioLabs: description from llminfo research: release date
Tags from its sources:instructtoolsreasoningvisionopen-sourcemoelong-contextcodingfunction-calling

Advantages and trade-offs

Compared with 10 chat models, picked mainly by popularity and IQ

Advantages

Open weights available

The weights are public. 2 of the 9 peers that say either way have open weights.

We only make a claim when there's enough data on models of the same type. Updated 22 Nov, 04:34 UTC.

Input & output

as stated by the model's providers
Text
Images
Video
Files
Text

Capabilities

Supported operations and provider-stated features

VisionReasoningFunction callingStructured outputPrompt cachingWeb search

Listings

 Moonshot AImoonshot/kimi-k3.1-proSources: LiteLLM data, Models.dev data1M$2.00cache read $0.20$10.00cache write 5m $2.00cache write 1h $4.00——
 Hugging FaceInference by Relacemoonshotai/Kimi-K3.1-ProSource: Hugging Face Inference Providers data1M$0.2403$8.4023——
 OpenRouterInference by Relacemoonshotai/kimi-k3.1-pro · fp4Source: OpenRouter data docs1M$0.2403cache read $0.20$8.4023241,241ms
 Nous Portalmoonshotai/kimi-k3.1-pro:batch · batchSource: Nous Portal Inference Gateway data1M$0.372cache read $0.10$5.583——
 Flowdotaccounts/fireworks/models/kimi-k3p1-proSource: FlowDot data1M$0.50$0.50——
 Kenarikimi-k3-1-proSource: Models.dev data1M$0.80cache read $0.08$8.00——
 Aihubmixcoding-kimi-k3.1-pro · codingSource: AIHubMix data docsNot published by the provider$0.88cache read $0.132$3.23——
 Jzstokenkimi-k3.1-proSource: Models.dev data1M$1.17cache read $0.0778$5.84cache write $1.17——
 Inferriskimi-k3.1-proSource: Models.dev data1M$1.20cache read $0.12$6.00——
 Neuralwattkimi-k3.1-pro-flex · flexSources: Models.dev data, NeuralWatt data1M$1.30cache read $0.13$6.50——

18 listings without a price are hidden.

Included in plans

CheapestInference Flagship PoolcheapestinferenceSubscription$199/moPlan page
Umans Code ProumansSubscription$20/moslows down past the allowancePlan page
Umans Code MaxumansSubscription$50/moslows down past the allowancePlan page

Benchmarks

Named effort levels run from lowest to highest. Select a result to see its source details.

Human preference

Arena ELO39 related benchmarks · 39 results
Design Arena EloSource: Design Arena Leaderboards data · Nov 2026

Agentic

Arena AgentSource: LMArena data · Nov 2026
Design Arena EloSource: Design Arena Leaderboards data · Nov 2026
Source: OpenRouter benchmark exports · Nov 2026
23 Oct 2026 version · 90-day results
Across 20 providers lowest 72.7% median 75.6% highest 78.7%
Show results from 20 providersHide results from 20 providers
  • DigitalOcean
  • Fireworks
  • Relace
  • Phala
  • Moonshot AI
  • Decart
  • Morph
  • InferenceNet
  • Wafer
  • Nebius
  • River
  • AkashML
  • Alibaba
  • DeepInfra
  • PrimaLabs
  • Sail Research
  • Parasail
  • Together
  • Modal
  • Makora
ValsSource: Vals AI

Science

GPQASource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
GPQA Diamond20 results
Source: OpenRouter benchmark exports · Nov 2026
23 Oct 2026 version · 90-day results
Across 19 providers lowest 91.6% median 92.9% highest 93.9%OpenRouter's automatic routing 93.6%
Show results from 19 providers and OpenRouter's automatic routingHide results from 19 providers and OpenRouter's automatic routing
  • Sail Research
  • Nebius
  • River
  • OpenRouter's automatic routing
  • Decart
  • DigitalOcean
  • Modal
  • Parasail
  • InferenceNet
  • Moonshot AI
  • Morph
  • AkashML
  • Fireworks
  • PrimaLabs
  • Phala
  • Together
  • Wafer
  • Makora
  • Relace
  • DeepInfra
ValsSource: Vals AI

Coding

Arena ELOSource: LMArena data · Nov 2026
BenchCADSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
DeepSWESource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim

Reasoning

ZEROBenchSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim

General

Vision

Instruction following

Math

Other

SuperCLUESource: SuperCLUE data
ValsSource: Vals AI

Data credits: Data by LLM Stats · Hugging Face official benchmark leaderboards; licence varies by benchmark · Epoch AI Benchmarking Hub, CC BY 4.0 (licence) · LMArena, CC BY 4.0 (licence) · Baromètre de l’écriture IA, Johannes Eva; public numeric ranking data, © 2026 Johannes Eva, no results licence stated · BullshitBench by Peter Gostev, MIT (licence) · EQ-Bench (Samuel J. Paech) — cite paech2023eqbench · OpenRouter public benchmark leaderboard and search-lane exports, CC BY 4.0. Fractional source scores converted to percentages; immutable dated snapshots. (licence) · MineBench, MIT (licence) · Clinical Benchmarks (clinicalbenchmarks.ai), CC BY 4.0; fractions converted to percentages; source documents retain their own terms (licence) · WeirdML v3 by Håvard Tveit Ihle (Norwegian Defence Research Establishment). API costs supported by Epoch AI, METR and NDRE.

Around the web

Also known as: accounts/fireworks/models/kimi-k3p1-pro, alibaba/kimi-k3.1-pro, azure_ai/kimi-k3.1-pro, baseten/moonshotai/kimi-k3.1-pro, cline-pass/kimi-k3.1-pro, coding-kimi-k3.1-pro, coding-kimi-k3.1-pro-highspeed, databricks/databricks-kimi-k3-1-pro, e2ee-kimi-k3-1-pro-p, empiriolabs/kimi-k3-1-pro, fireworks/kimi-k3.1-pro, fireworks_ai/accounts/fireworks/models/kimi-k3p1-pro, fireworks_ai/accounts/fireworks/routers/kimi-k3p1-pro-us, gmi/moonshotai/kimi-k3.1-pro, hf:moonshotai/kimi-k3.1-pro, kimi k3.1 pro, kimi k3.1 pro highspeed, kimi-k3-1-pro, kimi-k3-1-pro-highspeed, kimi-k3.1-pro-flex, kimi-k3.1-pro-highspeed, kimi-k3.1-pro-max, kimi-k3.1-pro-us, kimi-k3.1-pro@eu, moonshot/kimi-k3-1-pro, moonshot/kimi-k3-1-pro-highspeed, moonshot/kimi-k3.1-pro, moonshot/kimi-k3.1-pro-highspeed, moonshotai/kimi-k3.1-pro, moonshotai/kimi-k3.1-pro-highspeed, moonshotai/kimi-k3.1-pro:batch, nebius/moonshotai/kimi-k3.1-pro, opencode_go/kimi-k3.1-pro, parasail-kimi-k31-pro, phala/kimi-k3.1-pro, runpod/kimi-k3.1-pro, runpod/kimi-k3.1-pro-highspeed, runware/kimi-k3.1-pro, scx-ai-gp/kimi-k3.1-pro, scx-ai-gp/kimi-k3.1-pro-highspeed, tee/kimi-k3.1-pro, together-ai/kimi-k3.1-pro, together_ai/moonshotai/kimi-k3.1-pro, umans-kimi-k3.1-pro

Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works