Skip to content
ModelCap

Kimi K2 Thinking

Current board #321 Arena configurationsRestricted license

Moonshot AI·moonshotai/kimi-k2-thinking

Kimi K2 Thinking is a language model published by Moonshot AI. 2 OpenRouter endpoints are published in the latest snapshot. It accepts text input and returns text. The published context limit is 262K tokens. The catalogue declares support for reasoning, tool use, structured outputs.

ModelCap Index · Current board
70.8#32
Measured · 86% support · score interval 70.8–77.0
Market Gravity
33.6
Secondary market signal
Input
$0.60
per 1M tokens
Output
$2.50
per 1M tokens
Context
262K
tokens
Providers
2
endpoints
Gateway spend
—
latest · Vercel share

Providers

Uptime measured over the last 30 minutes
ProviderInputOutputContextQuantUptime
Novitabf16Cheapest$0.60$2.50262Kbf1693.99%
Google$0.60$2.50262K—98.94%

Reasoning efforts

Best supported result can inform ModelCap

Published Max, X-High, High and other configurations for this model. These are sourced scores, not head-to-head essays.

Related community configurations

Human preference rating
ConfigurationRatingSource rankVotesSource
Turbo
kimi-k2-thinking-turbo
1430
CI 1427–1433
#9362KArena
Match 98%

Reported evaluation leads (5)

Excluded from score

Structured rows reported in this exact model repository at the pinned revision below. They are useful research leads, but remain quarantined until dataset revision, harness, configuration and identity pass ModelCap’s independent admission review.

Dataset / taskMetricValueRevisionProvenance
cais/hle
hle
—44.9
Unpinned
not published
Reported
Quarantined
cais/hle
hle
—23.9
Unpinned
not published
Reported
Quarantined
harborframework/terminal-bench-2.0
terminalbench_2
—35.7
Unpinned
not published
Reported
Quarantined
Idavidrein/gpqa
diamond
—84.5
Unpinned
not published
Reported
Quarantined
SWE-bench/SWE-bench_Verified
swe_bench_%_resolved
—71.3
Unpinned
not published
Reported
Quarantined

ModelCap Index · Current board

Methodology
70.8
Measured · 3 public benchmark observations across 3 boards
Current board
#32
Public rank basis
Measured (measured)
Index support
86.4% · strong
Index interval
70.8–77.0
Rank posterior
Not available
Top 5 / Top 10 probability
Not available / Not available
Posterior as of
Snapshot timestamp unavailable
Identity binding
Exact catalogue product · moonshotai/moonshotai/kimi-k2-thinking · aggregates disclosed benchmark configurations · endpoint configuration not claimed · artifact metadata moonshotai/Kimi-K2-Thinking@a51ccc050d73dab088bf7b0e2dd9b30ae85a4e55 (not the evaluation revision)
Direct benchmark observations
3
Breadth-adjusted BES point estimate
73.9
Bayesian Evidence Score · measured evidence input only

BES normalizes admitted public benchmark evidence for measured Index rows. Its legacy score and observables prior do not define the public language rank.

Observed capability
73.9
Evidence status
confirmed · 84% mass
General preference
75.7
Coding
67.0
Agents & tools
—
Reasoning
—
Evidence breadth
92%
Evidence coverage
86%
Market and catalogue signals · secondary, never the language Index rank
Usage (OpenRouter popularity)
10.0
Liquidity (providers × uptime)
75.3
Open reach (HF downloads)
—
Surface (context / tools / modalities)
75.3
Freshness
35.2
  • · OpenRouter popularity #169
  • · 2 live providers
  • · Catalogue: context 262144, reasoning, tools
  • · First seen 2025-11-06

These adoption and deployment observations remain context only. They do not change this model's ModelCap Index score or rank.

BES evidence selected · measured input only

The bounded 0–100 capability score blends each source’s competitive placement with its published achievement, then combines capability families. Evidence breadth adds a modest uncertainty adjustment; price and popularity are not part of benchmark-led rank.

Deployment readiness

Metadata index
79.4
strong · 89% metadata coverage
Not capability
Artifact reproducibility
85.0
Access & legal clarity
67.5
Deployability
89.3
Evaluation provenance
68.1

Missing: artifactReproducibility.base-lineage, accessLegalClarity.resolvable-license-terms, evaluationProvenance.evaluation-dataset-revisions.

A metadata completeness and deployability index—not a safety certification, quality grade, or production-readiness claim.

Market activity

Venue coverage
Market Gravity
33.6
OpenRouter weekly popularity
#169 of 298

Market Gravity uses OpenRouter’s full-catalogue popularity order. Sparse Vercel top-ten observations appear only when published and do not affect the score.

Arena preference

Configuration results available

Arena has published current community results for Turbo. No default-configuration result is published, so the raw Arena field stays configuration-specific. The strongest supported result can still inform ModelCap Score with reduced confidence.

Current board through 6 Aug 2026, 13:00 UTC

Market Gravity breakdown

Usage55%
10.0

OpenRouter weekly popularity across the full model catalogue.

Liquidity25%
75.3

Independent providers versus the model's open or closed cohort, weighted by uptime.

Open reach15%
50.0

Hugging Face 30-day downloads for open models; neutral for closed models.

Freshness5%
35.2

Time since first public availability, on a six-month half-life.

Specification

Context window
262K tokens
Max output
100K tokens
Inputs
Text
Outputs
Text
Tokenizer
Other
Cached input tokens
$0.15 / 1M
First seen on OpenRouter
6 Nov 2025

Capabilities

  • Supported: Reasoning
  • Supported: Tool use
  • Supported: Structured output
  • Supported: Response format
  • Not supported: Moderated

Weights & access

Restricted license. Downloadable weights use a model-specific or use-restricted license. Review the linked license before use, redistribution, or commercial deployment.

Access
Restricted license
Licence
modified-mit
Pinned revision
a51ccc050d73
Parameters
1058.1B
Architecture
DeepseekV3ForCausalLM
Model type
kimi_k2
Hugging Face downloads (30d)
45K
Downloads (all time)
1.8M
Likes
2K
Repository
moonshotai/Kimi-K2-Thinking
Repository updated
30 Jan 2026, 04:49 UTC