VENICE ATLAS · MODEL DOSSIER · KIMI K2.5

Model overview

Kimi K2.5

Name: Kimi K2.5
Brand: venice.ai
Price: 0.5600 USD
Availability: InStock

Kimi K2.5 is Moonshot AIs most advanced open reasoning model, featuring trillion-parameter Mixture-of-Experts architecture with 32B active parameters and 256K context windows.

kimi-k2-5 · venice.ai

Text · Venice catalog$0.56 per 1M tokensPrice on Venice AI

ModalityTextOn Venice AI

Primary price$0.56 per 1M tokensper 1M tokens

Context window256KMax tokens

Benchmarks40Benchmark results

Benchmark results

See the tests behind each score. They use the benchmark’s original scoring scale.

Biology#2332.45 rating Chemistry#2632.65 rating Vision#4027.35 rating Finance#831.36 rating Frontend Development#2122.39 rating Healthcare#635.97 rating Image To Text#227.47 rating Language#931.36 rating Legal#931.36 rating Long Context#1024.48 rating Math#2736.67 rating Multimodal#3228.94 rating Physics#2732.65 rating Safety#22-6.29 rating Search#1622.27 rating Structured Output#1222.31 rating Video understanding#820.67 rating

Coding3 results

CyberGymcybergym · code

41.3%

ReportedOct 2026

FrontierSWEfrontierswe · code

26.0%

ReportedOct 2026

PaperBenchpaperbench · code

63.5%

ReportedOct 2026

Finance1 result

FinSearchComp T2&T3finsearchcomp-t2-t3 · finance

67.8%

ReportedOct 2026

Language1 result

MMLU-Prommlu-pro · language

87.1%

ReportedOct 2026

Long Context1 result

LongBench v2longbench-v2 · long_context

61.0%

ReportedOct 2026

Math6 results

AIME 2025aime-2025 · math

96.1%

ReportedOct 2026

HMMT 2025hmmt-2025 · math

95.4%

ReportedOct 2026

IMO-AnswerBenchimo-answerbench · math

81.8%

ReportedOct 2026

LiveBenchlivebench · math

69.1%

ReportedOct 2026

MathVisionmathvision · math

84.2%

ReportedOct 2026

MathVista-Minimathvista-mini · math

90.1%

ReportedOct 2026

Multimodal2 results

Video-MMEvideo-mme · multimodal

87.4%

ReportedOct 2026

VideoMMMUvideommmu · multimodal

86.6%

ReportedOct 2026

Reasoning13 results

AA-LCRaa-lcr · reasoning

70.0%

ReportedOct 2026

BrowseCompbrowsecomp · reasoning

74.9%

ReportedOct 2026

CharXiv-Rcharxiv-r · reasoning

77.5%

ReportedOct 2026

GPQAgpqa · reasoning

87.6%

ReportedOct 2026

Humanity's Last Examhumanity's-last-exam · reasoning

50.2%

ReportedOct 2026

LiveCodeBench v6livecodebench-v6 · reasoning

85.0%

ReportedOct 2026

OJBench (C++)ojbench-cpp · reasoning

57.4%

ReportedOct 2026

SciCodescicode · reasoning

48.7%

ReportedOct 2026

Seal-0seal-0 · reasoning

57.4%

ReportedOct 2026

SWE-bench Multilingualswe-bench-multilingual · reasoning

73.0%

ReportedOct 2026

SWE-Bench Proswe-bench-pro · reasoning

50.7%

ReportedOct 2026

SWE-Bench Verifiedswe-bench-verified · reasoning

76.8%

ReportedOct 2026

Terminal-Bench 2.0terminal-bench-2 · reasoning

50.8%

ReportedOct 2026

Search2 results

DeepSearchQAdeepsearchqa · search

77.1%

ReportedOct 2026

WideSearchwidesearch · search

79.0%

ReportedOct 2026

Vision1 result

InfoVQAtestinfovqatest · vision

92.6%

ReportedOct 2026

Learn what each benchmark measures ↗

Pricing detail

Input price per 1M tokens: $0.56
Output price per 1M tokens: $3.50
Cache read price per 1M tokens: $0.22
Blended price per 1M tokens (50/50): $2.03

Prices are published by Venice AI. Venice model page ↗

Capabilities

ReasoningFunction callingVisionWeb search

Endpoint metadata

Privacy: private
Quantization: not-available
Maximum completion tokens: 66K
Uncensored: No
Beta: No
Mapping confidence: 100.0%

View model source ↗

All capabilities

Max Images: 10
Quantization: not-available
Supports E2 E E: No
Supports Vision: Yes
Supports X Search: No
Optimized For Code: Yes
Supports Log Probs: Yes
Supports Reasoning: Yes
Supports Web Search: Yes
Supports Audio Input: No
Supports Video Input: No
Default Reasoning Effort: high
Reasoning Effort Options: none, low, medium, high
Supports Multiple Images: Yes
Supports Response Schema: Yes
Supports Tee Attestation: No
Supports Function Calling: Yes
Supports Reasoning Effort: Yes

Related models

More from moonshotai

kimi-k2-6Kimi K2.6$0.75 input per 1M tokens ↗kimi-k2-7-codeKimi K2.7 Code$0.75 input per 1M tokens ↗kimi-k3Kimi K3$3.75 input per 1M tokens ↗