MoonshotAI: Kimi K2 0711

Kimi K2 Instruct is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass. It is optimized for …

moonshotai/kimi-k2

Kimi K2 Instruct is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass. It is optimized for agentic capabilities, including advanced tool use, reasoning, and code synthesis. Kimi K2 excels across a broad range of benchmarks, particularly in coding (LiveCodeBench, SWE-bench), reasoning (ZebraLogic, GPQA), and tool-use (Tau2, AceBench) tasks. It supports long-context inference up to 128K tokens and is designed with a novel training stack that includes the MuonClip optimizer for stable large-scale MoE training.

Specifications & capabilities

Input
text
Output
text
Context window · tokens
131072
Knowledge cutoff
2024-12-31
Published model ID
Kimi-K2-Instruct
Capabilities
structuredOutputs, toolCalling
Description source ↗

Evaluations

21–40 / 60

LLM Stats · GitHub archive

0.784

source id
llmStatsArchive
leaderboard name
CSimpleQA · Raw
dimension
general
metric
CSimpleQA · Raw
rank
1
aggregation scope
status
archived
license
CC-BY-4.0
direction
higherIsBetter
conditions
language
en
benchmark
name
CSimpleQA
language
en
modality
text
verified
false
max_score
1
categories
  • general
  • language
created_at
2025-07-19T19:56:11.931358+00:00
paper_link
https://arxiv.org/abs/2411.07140
updated_at
2025-07-19T19:56:11.931358+00:00
description
Chinese SimpleQA is the first comprehensive Chinese benchmark to evaluate the factuality ability of language models to answer short questions. It contains 3,000 high-quality questions spanning 6 major topics with 99 diverse subtopics, designed to assess Chinese factual knowledge across humanities, science, engineering, culture, and society.
benchmark_id
csimpleqa
multilingual
true
analysis_method
Correct
is_self_reported
true
verified_by_llmstats
false
attribution
Jonathan Chavez Tamales and LLM Stats contributors (copyright 2024 jc); TensorTrace structures original results and derives scoped ranks.
license_url
https://creativecommons.org/licenses/by/4.0/
metric_kind
CSimpleQA · Raw
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
source_file
data/organizations/moonshotai/models/kimi-k2-instruct/benchmarks.json
archive_kind
llmStats
evidence_url
https://moonshotai.github.io/Kimi-K2/
git_revision
872b75f63b8d853c46ed17856424029a0c47b1d2
published_at
2025-10-24T17:47:59Z
snapshot_date
2025-10-24T17:47:59Z
source_revision
6e27aa41f51fa8cb327602349240077d1c7acae9fead8ba2f07db0d549927ddc
publisher_anchor
moonshotai/kimi-k2-instruct
metric_population
3
current_catalog_url
https://llm-stats.com/
benchmark_evaluation
true
source model id
kimi-k2-instruct
Original result ↗

LLM Stats · GitHub archive

0.784

source id
llmStatsArchive
leaderboard name
CSimpleQA · Normalized (reported)
dimension
general
metric
CSimpleQA · Normalized (reported)
rank
1
aggregation scope
status
archived
license
CC-BY-4.0
direction
higherIsBetter
conditions
language
en
benchmark
name
CSimpleQA
language
en
modality
text
verified
false
max_score
1
categories
  • general
  • language
created_at
2025-07-19T19:56:11.931358+00:00
paper_link
https://arxiv.org/abs/2411.07140
updated_at
2025-07-19T19:56:11.931358+00:00
description
Chinese SimpleQA is the first comprehensive Chinese benchmark to evaluate the factuality ability of language models to answer short questions. It contains 3,000 high-quality questions spanning 6 major topics with 99 diverse subtopics, designed to assess Chinese factual knowledge across humanities, science, engineering, culture, and society.
benchmark_id
csimpleqa
multilingual
true
analysis_method
Correct
is_self_reported
true
verified_by_llmstats
false
attribution
Jonathan Chavez Tamales and LLM Stats contributors (copyright 2024 jc); TensorTrace structures original results and derives scoped ranks.
license_url
https://creativecommons.org/licenses/by/4.0/
metric_kind
CSimpleQA · Normalized (reported)
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
source_file
data/organizations/moonshotai/models/kimi-k2-instruct/benchmarks.json
archive_kind
llmStats
evidence_url
https://moonshotai.github.io/Kimi-K2/
git_revision
872b75f63b8d853c46ed17856424029a0c47b1d2
published_at
2025-10-24T17:47:59Z
snapshot_date
2025-10-24T17:47:59Z
source_revision
6e27aa41f51fa8cb327602349240077d1c7acae9fead8ba2f07db0d549927ddc
publisher_anchor
moonshotai/kimi-k2-instruct
metric_population
3
current_catalog_url
https://llm-stats.com/
benchmark_evaluation
true
source model id
kimi-k2-instruct
Original result ↗

LLM Stats · GitHub archive

0.765

source id
llmStatsArchive
leaderboard name
ACEBench · Normalized (reported)
dimension
general
metric
ACEBench · Normalized (reported)
rank
1
aggregation scope
status
archived
license
CC-BY-4.0
direction
higherIsBetter
conditions
language
en
benchmark
name
ACEBench
language
en
modality
text
verified
false
max_score
1
categories
  • general
  • reasoning
created_at
2025-09-05T00:00:00.000000+00:00
paper_link
https://arxiv.org/abs/2501.12851
updated_at
2025-09-30T00:00:00.000000+00:00
description
ACEBench is a comprehensive benchmark for evaluating Large Language Models' tool usage capabilities across three primary evaluation types: Normal (basic tool usage scenarios), Special (tool usage with ambiguous or incomplete instructions), and Agent (multi-agent interactions simulating real-world dialogues). The benchmark covers 4,538 APIs across 8 major domains and 68 sub-domains including technology, finance, entertainment, society, health, culture, and environment, supporting both English and Chinese languages.
benchmark_id
acebench
multilingual
false
implementation_link
https://github.com/ACEBench/ACEBench
analysis_method
Accuracy
is_self_reported
true
verified_by_llmstats
false
attribution
Jonathan Chavez Tamales and LLM Stats contributors (copyright 2024 jc); TensorTrace structures original results and derives scoped ranks.
license_url
https://creativecommons.org/licenses/by/4.0/
metric_kind
ACEBench · Normalized (reported)
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
source_file
data/organizations/moonshotai/models/kimi-k2-instruct/benchmarks.json
archive_kind
llmStats
evidence_url
https://moonshotai.github.io/Kimi-K2/
git_revision
872b75f63b8d853c46ed17856424029a0c47b1d2
published_at
2025-10-24T17:47:59Z
snapshot_date
2025-10-24T17:47:59Z
source_revision
6e27aa41f51fa8cb327602349240077d1c7acae9fead8ba2f07db0d549927ddc
publisher_anchor
moonshotai/kimi-k2-instruct
metric_population
2
current_catalog_url
https://llm-stats.com/
benchmark_evaluation
true
source model id
kimi-k2-instruct
Original result ↗

LLM Stats · GitHub archive

0.765

source id
llmStatsArchive
leaderboard name
ACEBench · Raw
dimension
general
metric
ACEBench · Raw
rank
1
aggregation scope
status
archived
license
CC-BY-4.0
direction
higherIsBetter
conditions
language
en
benchmark
name
ACEBench
language
en
modality
text
verified
false
max_score
1
categories
  • general
  • reasoning
created_at
2025-09-05T00:00:00.000000+00:00
paper_link
https://arxiv.org/abs/2501.12851
updated_at
2025-09-30T00:00:00.000000+00:00
description
ACEBench is a comprehensive benchmark for evaluating Large Language Models' tool usage capabilities across three primary evaluation types: Normal (basic tool usage scenarios), Special (tool usage with ambiguous or incomplete instructions), and Agent (multi-agent interactions simulating real-world dialogues). The benchmark covers 4,538 APIs across 8 major domains and 68 sub-domains including technology, finance, entertainment, society, health, culture, and environment, supporting both English and Chinese languages.
benchmark_id
acebench
multilingual
false
implementation_link
https://github.com/ACEBench/ACEBench
analysis_method
Accuracy
is_self_reported
true
verified_by_llmstats
false
attribution
Jonathan Chavez Tamales and LLM Stats contributors (copyright 2024 jc); TensorTrace structures original results and derives scoped ranks.
license_url
https://creativecommons.org/licenses/by/4.0/
metric_kind
ACEBench · Raw
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
source_file
data/organizations/moonshotai/models/kimi-k2-instruct/benchmarks.json
archive_kind
llmStats
evidence_url
https://moonshotai.github.io/Kimi-K2/
git_revision
872b75f63b8d853c46ed17856424029a0c47b1d2
published_at
2025-10-24T17:47:59Z
snapshot_date
2025-10-24T17:47:59Z
source_revision
6e27aa41f51fa8cb327602349240077d1c7acae9fead8ba2f07db0d549927ddc
publisher_anchor
moonshotai/kimi-k2-instruct
metric_population
2
current_catalog_url
https://llm-stats.com/
benchmark_evaluation
true
source model id
kimi-k2-instruct
Original result ↗

LLM Stats · GitHub archive

0.764

source id
llmStatsArchive
leaderboard name
LiveBench · Raw
dimension
general
metric
LiveBench · Raw
rank
1
aggregation scope
status
archived
license
CC-BY-4.0
direction
higherIsBetter
conditions
language
en
benchmark
name
LiveBench
language
en
modality
text
verified
false
max_score
1
categories
  • math
  • reasoning
  • general
created_at
2025-09-05T00:00:00.000000+00:00
paper_link
https://arxiv.org/abs/2406.19314
updated_at
2025-09-05T00:00:00.000000+00:00
description
LiveBench is a challenging, contamination-limited LLM benchmark that addresses test set contamination by releasing new questions monthly based on recently-released datasets, arXiv papers, news articles, and IMDb movie synopses. It comprises tasks across math, coding, reasoning, language, instruction following, and data analysis with verifiable, objective ground-truth answers.
benchmark_id
livebench
multilingual
false
analysis_method
Pass@1
is_self_reported
true
verified_by_llmstats
false
attribution
Jonathan Chavez Tamales and LLM Stats contributors (copyright 2024 jc); TensorTrace structures original results and derives scoped ranks.
license_url
https://creativecommons.org/licenses/by/4.0/
metric_kind
LiveBench · Raw
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
source_file
data/organizations/moonshotai/models/kimi-k2-instruct/benchmarks.json
archive_kind
llmStats
evidence_url
https://moonshotai.github.io/Kimi-K2/
git_revision
872b75f63b8d853c46ed17856424029a0c47b1d2
published_at
2025-10-24T17:47:59Z
snapshot_date
2025-10-24T17:47:59Z
source_revision
6e27aa41f51fa8cb327602349240077d1c7acae9fead8ba2f07db0d549927ddc
publisher_anchor
moonshotai/kimi-k2-instruct
metric_population
1
current_catalog_url
https://llm-stats.com/
benchmark_evaluation
true
source model id
kimi-k2-instruct
Original result ↗

LLM Stats · GitHub archive

0.764

source id
llmStatsArchive
leaderboard name
MuSR · Normalized (reported)
dimension
general
metric
MuSR · Normalized (reported)
rank
1
aggregation scope
status
archived
license
CC-BY-4.0
direction
higherIsBetter
conditions
language
en
benchmark
name
MuSR
language
en
modality
text
verified
false
max_score
1
categories
  • reasoning
created_at
2025-07-19T19:56:12.708705+00:00
paper_link
https://arxiv.org/abs/2310.16049
updated_at
2025-07-19T19:56:12.708705+00:00
description
MuSR (Multistep Soft Reasoning) is a benchmark for evaluating language models on multistep soft reasoning tasks specified in natural language narratives. Created through a neurosymbolic synthetic-to-natural generation algorithm, it generates complex reasoning scenarios like murder mysteries roughly 1000 words in length that challenge current LLMs including GPT-4. The benchmark tests chain-of-thought reasoning capabilities across domains involving commonsense reasoning about physical and social situations.
benchmark_id
musr
multilingual
false
analysis_method
Pass@1
is_self_reported
true
verified_by_llmstats
false
attribution
Jonathan Chavez Tamales and LLM Stats contributors (copyright 2024 jc); TensorTrace structures original results and derives scoped ranks.
license_url
https://creativecommons.org/licenses/by/4.0/
metric_kind
MuSR · Normalized (reported)
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
source_file
data/organizations/moonshotai/models/kimi-k2-instruct/benchmarks.json
archive_kind
llmStats
evidence_url
https://moonshotai.github.io/Kimi-K2/
git_revision
872b75f63b8d853c46ed17856424029a0c47b1d2
published_at
2025-10-24T17:47:59Z
snapshot_date
2025-10-24T17:47:59Z
source_revision
6e27aa41f51fa8cb327602349240077d1c7acae9fead8ba2f07db0d549927ddc
publisher_anchor
moonshotai/kimi-k2-instruct
metric_population
1
current_catalog_url
https://llm-stats.com/
benchmark_evaluation
true
source model id
kimi-k2-instruct
Original result ↗

LLM Stats · GitHub archive

0.764

source id
llmStatsArchive
leaderboard name
LiveBench · Normalized (reported)
dimension
general
metric
LiveBench · Normalized (reported)
rank
1
aggregation scope
status
archived
license
CC-BY-4.0
direction
higherIsBetter
conditions
language
en
benchmark
name
LiveBench
language
en
modality
text
verified
false
max_score
1
categories
  • math
  • reasoning
  • general
created_at
2025-09-05T00:00:00.000000+00:00
paper_link
https://arxiv.org/abs/2406.19314
updated_at
2025-09-05T00:00:00.000000+00:00
description
LiveBench is a challenging, contamination-limited LLM benchmark that addresses test set contamination by releasing new questions monthly based on recently-released datasets, arXiv papers, news articles, and IMDb movie synopses. It comprises tasks across math, coding, reasoning, language, instruction following, and data analysis with verifiable, objective ground-truth answers.
benchmark_id
livebench
multilingual
false
analysis_method
Pass@1
is_self_reported
true
verified_by_llmstats
false
attribution
Jonathan Chavez Tamales and LLM Stats contributors (copyright 2024 jc); TensorTrace structures original results and derives scoped ranks.
license_url
https://creativecommons.org/licenses/by/4.0/
metric_kind
LiveBench · Normalized (reported)
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
source_file
data/organizations/moonshotai/models/kimi-k2-instruct/benchmarks.json
archive_kind
llmStats
evidence_url
https://moonshotai.github.io/Kimi-K2/
git_revision
872b75f63b8d853c46ed17856424029a0c47b1d2
published_at
2025-10-24T17:47:59Z
snapshot_date
2025-10-24T17:47:59Z
source_revision
6e27aa41f51fa8cb327602349240077d1c7acae9fead8ba2f07db0d549927ddc
publisher_anchor
moonshotai/kimi-k2-instruct
metric_population
1
current_catalog_url
https://llm-stats.com/
benchmark_evaluation
true
source model id
kimi-k2-instruct
Original result ↗

LLM Stats · GitHub archive

0.764

source id
llmStatsArchive
leaderboard name
MuSR · Raw
dimension
general
metric
MuSR · Raw
rank
1
aggregation scope
status
archived
license
CC-BY-4.0
direction
higherIsBetter
conditions
language
en
benchmark
name
MuSR
language
en
modality
text
verified
false
max_score
1
categories
  • reasoning
created_at
2025-07-19T19:56:12.708705+00:00
paper_link
https://arxiv.org/abs/2310.16049
updated_at
2025-07-19T19:56:12.708705+00:00
description
MuSR (Multistep Soft Reasoning) is a benchmark for evaluating language models on multistep soft reasoning tasks specified in natural language narratives. Created through a neurosymbolic synthetic-to-natural generation algorithm, it generates complex reasoning scenarios like murder mysteries roughly 1000 words in length that challenge current LLMs including GPT-4. The benchmark tests chain-of-thought reasoning capabilities across domains involving commonsense reasoning about physical and social situations.
benchmark_id
musr
multilingual
false
analysis_method
Pass@1
is_self_reported
true
verified_by_llmstats
false
attribution
Jonathan Chavez Tamales and LLM Stats contributors (copyright 2024 jc); TensorTrace structures original results and derives scoped ranks.
license_url
https://creativecommons.org/licenses/by/4.0/
metric_kind
MuSR · Raw
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
source_file
data/organizations/moonshotai/models/kimi-k2-instruct/benchmarks.json
archive_kind
llmStats
evidence_url
https://moonshotai.github.io/Kimi-K2/
git_revision
872b75f63b8d853c46ed17856424029a0c47b1d2
published_at
2025-10-24T17:47:59Z
snapshot_date
2025-10-24T17:47:59Z
source_revision
6e27aa41f51fa8cb327602349240077d1c7acae9fead8ba2f07db0d549927ddc
publisher_anchor
moonshotai/kimi-k2-instruct
metric_population
1
current_catalog_url
https://llm-stats.com/
benchmark_evaluation
true
source model id
kimi-k2-instruct
Original result ↗

LLM Stats · GitHub archive

0.6

source id
llmStatsArchive
leaderboard name
Aider-Polyglot · Normalized (reported)
dimension
general
metric
Aider-Polyglot · Normalized (reported)
rank
1
aggregation scope
status
archived
license
CC-BY-4.0
direction
higherIsBetter
conditions
language
en
benchmark
name
Aider-Polyglot
language
en
modality
text
verified
false
max_score
1
categories
  • general
  • code
created_at
2025-09-05T00:00:00.000000+00:00
updated_at
2025-09-30T00:00:00.000000+00:00
description
A coding benchmark that evaluates LLMs on 225 challenging Exercism programming exercises across C++, Go, Java, JavaScript, Python, and Rust. Models receive two attempts to solve each problem, with test error feedback provided after the first attempt if it fails. The benchmark measures both initial problem-solving ability and capacity to edit code based on error feedback, providing an end-to-end evaluation of code generation and editing capabilities across multiple programming languages.
benchmark_id
aider-polyglot
multilingual
false
implementation_link
https://github.com/Aider-AI/polyglot-benchmark
analysis_method
Accuracy
is_self_reported
true
verified_by_llmstats
false
attribution
Jonathan Chavez Tamales and LLM Stats contributors (copyright 2024 jc); TensorTrace structures original results and derives scoped ranks.
license_url
https://creativecommons.org/licenses/by/4.0/
metric_kind
Aider-Polyglot · Normalized (reported)
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
source_file
data/organizations/moonshotai/models/kimi-k2-instruct/benchmarks.json
archive_kind
llmStats
evidence_url
https://moonshotai.github.io/Kimi-K2/
git_revision
872b75f63b8d853c46ed17856424029a0c47b1d2
published_at
2025-10-24T17:47:59Z
snapshot_date
2025-10-24T17:47:59Z
source_revision
6e27aa41f51fa8cb327602349240077d1c7acae9fead8ba2f07db0d549927ddc
publisher_anchor
moonshotai/kimi-k2-instruct
metric_population
5
current_catalog_url
https://llm-stats.com/
benchmark_evaluation
true
source model id
kimi-k2-instruct
Original result ↗

LLM Stats · GitHub archive

0.6

source id
llmStatsArchive
leaderboard name
Aider-Polyglot · Raw
dimension
general
metric
Aider-Polyglot · Raw
rank
1
aggregation scope
status
archived
license
CC-BY-4.0
direction
higherIsBetter
conditions
language
en
benchmark
name
Aider-Polyglot
language
en
modality
text
verified
false
max_score
1
categories
  • general
  • code
created_at
2025-09-05T00:00:00.000000+00:00
updated_at
2025-09-30T00:00:00.000000+00:00
description
A coding benchmark that evaluates LLMs on 225 challenging Exercism programming exercises across C++, Go, Java, JavaScript, Python, and Rust. Models receive two attempts to solve each problem, with test error feedback provided after the first attempt if it fails. The benchmark measures both initial problem-solving ability and capacity to edit code based on error feedback, providing an end-to-end evaluation of code generation and editing capabilities across multiple programming languages.
benchmark_id
aider-polyglot
multilingual
false
implementation_link
https://github.com/Aider-AI/polyglot-benchmark
analysis_method
Accuracy
is_self_reported
true
verified_by_llmstats
false
attribution
Jonathan Chavez Tamales and LLM Stats contributors (copyright 2024 jc); TensorTrace structures original results and derives scoped ranks.
license_url
https://creativecommons.org/licenses/by/4.0/
metric_kind
Aider-Polyglot · Raw
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
source_file
data/organizations/moonshotai/models/kimi-k2-instruct/benchmarks.json
archive_kind
llmStats
evidence_url
https://moonshotai.github.io/Kimi-K2/
git_revision
872b75f63b8d853c46ed17856424029a0c47b1d2
published_at
2025-10-24T17:47:59Z
snapshot_date
2025-10-24T17:47:59Z
source_revision
6e27aa41f51fa8cb327602349240077d1c7acae9fead8ba2f07db0d549927ddc
publisher_anchor
moonshotai/kimi-k2-instruct
metric_population
5
current_catalog_url
https://llm-stats.com/
benchmark_evaluation
true
source model id
kimi-k2-instruct
Original result ↗

LLM Stats · GitHub archive

0.541

source id
llmStatsArchive
leaderboard name
Multi-Challenge · Raw
dimension
general
metric
Multi-Challenge · Raw
rank
1
aggregation scope
status
archived
license
CC-BY-4.0
direction
higherIsBetter
conditions
language
en
benchmark
name
Multi-Challenge
language
en
modality
text
verified
false
max_score
1
categories
  • communication
  • reasoning
created_at
2025-09-05T00:00:00.000000+00:00
paper_link
https://arxiv.org/abs/2501.17399
updated_at
2025-09-05T00:00:00.000000+00:00
description
MultiChallenge is a realistic multi-turn conversation evaluation benchmark that challenges frontier LLMs across four key categories: instruction retention (maintaining instructions throughout conversations), inference memory (recalling and connecting details from previous turns), reliable versioned editing (adapting to evolving instructions during collaborative editing), and self-coherence (avoiding contradictions in responses). The benchmark evaluates models on sustained, contextually complex dialogues across diverse topics including travel planning, technical documentation, and professional communication.
benchmark_id
multichallenge
multilingual
false
analysis_method
Accuracy
is_self_reported
true
verified_by_llmstats
false
attribution
Jonathan Chavez Tamales and LLM Stats contributors (copyright 2024 jc); TensorTrace structures original results and derives scoped ranks.
license_url
https://creativecommons.org/licenses/by/4.0/
metric_kind
Multi-Challenge · Raw
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
source_file
data/organizations/moonshotai/models/kimi-k2-instruct/benchmarks.json
archive_kind
llmStats
evidence_url
https://moonshotai.github.io/Kimi-K2/
git_revision
872b75f63b8d853c46ed17856424029a0c47b1d2
published_at
2025-10-24T17:47:59Z
snapshot_date
2025-10-24T17:47:59Z
source_revision
6e27aa41f51fa8cb327602349240077d1c7acae9fead8ba2f07db0d549927ddc
publisher_anchor
moonshotai/kimi-k2-instruct
metric_population
3
current_catalog_url
https://llm-stats.com/
benchmark_evaluation
true
source model id
kimi-k2-instruct
Original result ↗

LLM Stats · GitHub archive

0.541

source id
llmStatsArchive
leaderboard name
Multi-Challenge · Normalized (reported)
dimension
general
metric
Multi-Challenge · Normalized (reported)
rank
1
aggregation scope
status
archived
license
CC-BY-4.0
direction
higherIsBetter
conditions
language
en
benchmark
name
Multi-Challenge
language
en
modality
text
verified
false
max_score
1
categories
  • communication
  • reasoning
created_at
2025-09-05T00:00:00.000000+00:00
paper_link
https://arxiv.org/abs/2501.17399
updated_at
2025-09-05T00:00:00.000000+00:00
description
MultiChallenge is a realistic multi-turn conversation evaluation benchmark that challenges frontier LLMs across four key categories: instruction retention (maintaining instructions throughout conversations), inference memory (recalling and connecting details from previous turns), reliable versioned editing (adapting to evolving instructions during collaborative editing), and self-coherence (avoiding contradictions in responses). The benchmark evaluates models on sustained, contextually complex dialogues across diverse topics including travel planning, technical documentation, and professional communication.
benchmark_id
multichallenge
multilingual
false
analysis_method
Accuracy
is_self_reported
true
verified_by_llmstats
false
attribution
Jonathan Chavez Tamales and LLM Stats contributors (copyright 2024 jc); TensorTrace structures original results and derives scoped ranks.
license_url
https://creativecommons.org/licenses/by/4.0/
metric_kind
Multi-Challenge · Normalized (reported)
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
source_file
data/organizations/moonshotai/models/kimi-k2-instruct/benchmarks.json
archive_kind
llmStats
evidence_url
https://moonshotai.github.io/Kimi-K2/
git_revision
872b75f63b8d853c46ed17856424029a0c47b1d2
published_at
2025-10-24T17:47:59Z
snapshot_date
2025-10-24T17:47:59Z
source_revision
6e27aa41f51fa8cb327602349240077d1c7acae9fead8ba2f07db0d549927ddc
publisher_anchor
moonshotai/kimi-k2-instruct
metric_population
3
current_catalog_url
https://llm-stats.com/
benchmark_evaluation
true
source model id
kimi-k2-instruct
Original result ↗

LLM Stats · GitHub archive

0.537

source id
llmStatsArchive
leaderboard name
LiveCodeBench v6 · Normalized (reported)
dimension
general
metric
LiveCodeBench v6 · Normalized (reported)
rank
1
aggregation scope
status
archived
license
CC-BY-4.0
direction
higherIsBetter
conditions
language
en
benchmark
name
LiveCodeBench v6
language
en
modality
text
verified
false
max_score
1
categories
  • reasoning
  • general
created_at
2025-07-19T19:56:11.785682+00:00
paper_link
https://arxiv.org/abs/2403.07974
updated_at
2025-07-19T19:56:11.785682+00:00
description
LiveCodeBench is a holistic and contamination-free evaluation benchmark for large language models for code. It continuously collects new problems from programming contests (LeetCode, AtCoder, CodeForces) and evaluates four different scenarios: code generation, self-repair, code execution, and test output prediction. Problems are annotated with release dates to enable evaluation on unseen problems released after a model's training cutoff.
benchmark_id
livecodebench-v6
multilingual
false
analysis_method
Pass@1
is_self_reported
true
verified_by_llmstats
false
attribution
Jonathan Chavez Tamales and LLM Stats contributors (copyright 2024 jc); TensorTrace structures original results and derives scoped ranks.
license_url
https://creativecommons.org/licenses/by/4.0/
metric_kind
LiveCodeBench v6 · Normalized (reported)
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
source_file
data/organizations/moonshotai/models/kimi-k2-instruct/benchmarks.json
archive_kind
llmStats
evidence_url
https://moonshotai.github.io/Kimi-K2/
git_revision
872b75f63b8d853c46ed17856424029a0c47b1d2
published_at
2025-10-24T17:47:59Z
snapshot_date
2025-10-24T17:47:59Z
source_revision
6e27aa41f51fa8cb327602349240077d1c7acae9fead8ba2f07db0d549927ddc
publisher_anchor
moonshotai/kimi-k2-instruct
metric_population
2
current_catalog_url
https://llm-stats.com/
benchmark_evaluation
true
source model id
kimi-k2-instruct
Original result ↗

LLM Stats · GitHub archive

0.537

source id
llmStatsArchive
leaderboard name
LiveCodeBench v6 · Raw
dimension
general
metric
LiveCodeBench v6 · Raw
rank
1
aggregation scope
status
archived
license
CC-BY-4.0
direction
higherIsBetter
conditions
language
en
benchmark
name
LiveCodeBench v6
language
en
modality
text
verified
false
max_score
1
categories
  • reasoning
  • general
created_at
2025-07-19T19:56:11.785682+00:00
paper_link
https://arxiv.org/abs/2403.07974
updated_at
2025-07-19T19:56:11.785682+00:00
description
LiveCodeBench is a holistic and contamination-free evaluation benchmark for large language models for code. It continuously collects new problems from programming contests (LeetCode, AtCoder, CodeForces) and evaluates four different scenarios: code generation, self-repair, code execution, and test output prediction. Problems are annotated with release dates to enable evaluation on unseen problems released after a model's training cutoff.
benchmark_id
livecodebench-v6
multilingual
false
analysis_method
Pass@1
is_self_reported
true
verified_by_llmstats
false
attribution
Jonathan Chavez Tamales and LLM Stats contributors (copyright 2024 jc); TensorTrace structures original results and derives scoped ranks.
license_url
https://creativecommons.org/licenses/by/4.0/
metric_kind
LiveCodeBench v6 · Raw
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
source_file
data/organizations/moonshotai/models/kimi-k2-instruct/benchmarks.json
archive_kind
llmStats
evidence_url
https://moonshotai.github.io/Kimi-K2/
git_revision
872b75f63b8d853c46ed17856424029a0c47b1d2
published_at
2025-10-24T17:47:59Z
snapshot_date
2025-10-24T17:47:59Z
source_revision
6e27aa41f51fa8cb327602349240077d1c7acae9fead8ba2f07db0d549927ddc
publisher_anchor
moonshotai/kimi-k2-instruct
metric_population
2
current_catalog_url
https://llm-stats.com/
benchmark_evaluation
true
source model id
kimi-k2-instruct
Original result ↗

LLM Stats · GitHub archive

0.271

source id
llmStatsArchive
leaderboard name
OJBench · Normalized (reported)
dimension
general
metric
OJBench · Normalized (reported)
rank
1
aggregation scope
status
archived
license
CC-BY-4.0
direction
higherIsBetter
conditions
language
en
benchmark
name
OJBench
language
en
modality
text
verified
false
max_score
1
categories
  • reasoning
created_at
2025-09-05T00:00:00.000000+00:00
paper_link
https://arxiv.org/abs/2506.16395
updated_at
2025-09-05T00:00:00.000000+00:00
description
OJBench is a competition-level code benchmark designed to assess the competitive-level code reasoning abilities of large language models. It comprises 232 programming competition problems from NOI and ICPC, categorized into Easy, Medium, and Hard difficulty levels. The benchmark evaluates models' ability to solve complex competitive programming challenges using Python and C++.
benchmark_id
ojbench
multilingual
false
analysis_method
Pass@1
is_self_reported
true
verified_by_llmstats
false
attribution
Jonathan Chavez Tamales and LLM Stats contributors (copyright 2024 jc); TensorTrace structures original results and derives scoped ranks.
license_url
https://creativecommons.org/licenses/by/4.0/
metric_kind
OJBench · Normalized (reported)
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
source_file
data/organizations/moonshotai/models/kimi-k2-instruct/benchmarks.json
archive_kind
llmStats
evidence_url
https://moonshotai.github.io/Kimi-K2/
git_revision
872b75f63b8d853c46ed17856424029a0c47b1d2
published_at
2025-10-24T17:47:59Z
snapshot_date
2025-10-24T17:47:59Z
source_revision
6e27aa41f51fa8cb327602349240077d1c7acae9fead8ba2f07db0d549927ddc
publisher_anchor
moonshotai/kimi-k2-instruct
metric_population
2
current_catalog_url
https://llm-stats.com/
benchmark_evaluation
true
source model id
kimi-k2-instruct
Original result ↗

LLM Stats · GitHub archive

0.271

source id
llmStatsArchive
leaderboard name
OJBench · Raw
dimension
general
metric
OJBench · Raw
rank
1
aggregation scope
status
archived
license
CC-BY-4.0
direction
higherIsBetter
conditions
language
en
benchmark
name
OJBench
language
en
modality
text
verified
false
max_score
1
categories
  • reasoning
created_at
2025-09-05T00:00:00.000000+00:00
paper_link
https://arxiv.org/abs/2506.16395
updated_at
2025-09-05T00:00:00.000000+00:00
description
OJBench is a competition-level code benchmark designed to assess the competitive-level code reasoning abilities of large language models. It comprises 232 programming competition problems from NOI and ICPC, categorized into Easy, Medium, and Hard difficulty levels. The benchmark evaluates models' ability to solve complex competitive programming challenges using Python and C++.
benchmark_id
ojbench
multilingual
false
analysis_method
Pass@1
is_self_reported
true
verified_by_llmstats
false
attribution
Jonathan Chavez Tamales and LLM Stats contributors (copyright 2024 jc); TensorTrace structures original results and derives scoped ranks.
license_url
https://creativecommons.org/licenses/by/4.0/
metric_kind
OJBench · Raw
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
source_file
data/organizations/moonshotai/models/kimi-k2-instruct/benchmarks.json
archive_kind
llmStats
evidence_url
https://moonshotai.github.io/Kimi-K2/
git_revision
872b75f63b8d853c46ed17856424029a0c47b1d2
published_at
2025-10-24T17:47:59Z
snapshot_date
2025-10-24T17:47:59Z
source_revision
6e27aa41f51fa8cb327602349240077d1c7acae9fead8ba2f07db0d549927ddc
publisher_anchor
moonshotai/kimi-k2-instruct
metric_population
2
current_catalog_url
https://llm-stats.com/
benchmark_evaluation
true
source model id
kimi-k2-instruct
Original result ↗

LLM Stats · GitHub archive

0.25

source id
llmStatsArchive
leaderboard name
Terminus · Raw
dimension
general
metric
Terminus · Raw
rank
1
aggregation scope
status
archived
license
CC-BY-4.0
direction
higherIsBetter
conditions
language
en
benchmark
name
Terminus
language
en
modality
text
verified
false
max_score
1
categories
  • reasoning
  • code
created_at
2025-07-19T19:56:12.355994+00:00
paper_link
https://github.com/laude-institute/terminal-bench
updated_at
2025-07-19T19:56:12.355994+00:00
description
Terminal-Bench is a benchmark for testing AI agents in real terminal environments, evaluating how well agents can handle real-world, end-to-end tasks autonomously. The benchmark includes tasks spanning coding, system administration, security, data science, model training, file operations, version control, and web development. Terminus is the neutral test-bed agent designed to work with Terminal-Bench, operating purely through tmux sessions without dedicated tools.
benchmark_id
terminus
multilingual
false
analysis_method
Accuracy
is_self_reported
true
verified_by_llmstats
false
attribution
Jonathan Chavez Tamales and LLM Stats contributors (copyright 2024 jc); TensorTrace structures original results and derives scoped ranks.
license_url
https://creativecommons.org/licenses/by/4.0/
metric_kind
Terminus · Raw
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
source_file
data/organizations/moonshotai/models/kimi-k2-instruct/benchmarks.json
archive_kind
llmStats
evidence_url
https://moonshotai.github.io/Kimi-K2/
git_revision
872b75f63b8d853c46ed17856424029a0c47b1d2
published_at
2025-10-24T17:47:59Z
snapshot_date
2025-10-24T17:47:59Z
source_revision
6e27aa41f51fa8cb327602349240077d1c7acae9fead8ba2f07db0d549927ddc
publisher_anchor
moonshotai/kimi-k2-instruct
metric_population
1
current_catalog_url
https://llm-stats.com/
benchmark_evaluation
true
source model id
kimi-k2-instruct
Original result ↗

LLM Stats · GitHub archive

0.25

source id
llmStatsArchive
leaderboard name
Terminus · Normalized (reported)
dimension
general
metric
Terminus · Normalized (reported)
rank
1
aggregation scope
status
archived
license
CC-BY-4.0
direction
higherIsBetter
conditions
language
en
benchmark
name
Terminus
language
en
modality
text
verified
false
max_score
1
categories
  • reasoning
  • code
created_at
2025-07-19T19:56:12.355994+00:00
paper_link
https://github.com/laude-institute/terminal-bench
updated_at
2025-07-19T19:56:12.355994+00:00
description
Terminal-Bench is a benchmark for testing AI agents in real terminal environments, evaluating how well agents can handle real-world, end-to-end tasks autonomously. The benchmark includes tasks spanning coding, system administration, security, data science, model training, file operations, version control, and web development. Terminus is the neutral test-bed agent designed to work with Terminal-Bench, operating purely through tmux sessions without dedicated tools.
benchmark_id
terminus
multilingual
false
analysis_method
Accuracy
is_self_reported
true
verified_by_llmstats
false
attribution
Jonathan Chavez Tamales and LLM Stats contributors (copyright 2024 jc); TensorTrace structures original results and derives scoped ranks.
license_url
https://creativecommons.org/licenses/by/4.0/
metric_kind
Terminus · Normalized (reported)
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
source_file
data/organizations/moonshotai/models/kimi-k2-instruct/benchmarks.json
archive_kind
llmStats
evidence_url
https://moonshotai.github.io/Kimi-K2/
git_revision
872b75f63b8d853c46ed17856424029a0c47b1d2
published_at
2025-10-24T17:47:59Z
snapshot_date
2025-10-24T17:47:59Z
source_revision
6e27aa41f51fa8cb327602349240077d1c7acae9fead8ba2f07db0d549927ddc
publisher_anchor
moonshotai/kimi-k2-instruct
metric_population
1
current_catalog_url
https://llm-stats.com/
benchmark_evaluation
true
source model id
kimi-k2-instruct
Original result ↗

LLM Stats · GitHub archive

0.047

source id
llmStatsArchive
leaderboard name
Humanity's Last Exam · Raw
dimension
general
metric
Humanity's Last Exam · Raw
rank
1
aggregation scope
status
archived
license
CC-BY-4.0
direction
higherIsBetter
conditions
language
en
benchmark
name
Humanity's Last Exam
language
en
modality
multimodal
verified
false
max_score
1
categories
  • general
created_at
2025-07-19T19:56:12.507693+00:00
paper_link
https://arxiv.org/abs/2501.14249
updated_at
2025-07-19T19:56:12.507693+00:00
description
A multi-modal benchmark at the frontier of human knowledge with 2,500 questions across dozens of subjects including mathematics, humanities, and natural sciences, created by nearly 1000 subject expert contributors from over 500 institutions
benchmark_id
humanity's-last-exam
multilingual
false
analysis_method
Accuracy (Text Only)
is_self_reported
true
verified_by_llmstats
false
attribution
Jonathan Chavez Tamales and LLM Stats contributors (copyright 2024 jc); TensorTrace structures original results and derives scoped ranks.
license_url
https://creativecommons.org/licenses/by/4.0/
metric_kind
Humanity's Last Exam · Raw
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
source_file
data/organizations/moonshotai/models/kimi-k2-instruct/benchmarks.json
archive_kind
llmStats
evidence_url
https://moonshotai.github.io/Kimi-K2/
git_revision
872b75f63b8d853c46ed17856424029a0c47b1d2
published_at
2025-10-24T17:47:59Z
snapshot_date
2025-10-24T17:47:59Z
source_revision
6e27aa41f51fa8cb327602349240077d1c7acae9fead8ba2f07db0d549927ddc
publisher_anchor
moonshotai/kimi-k2-instruct
metric_population
1
current_catalog_url
https://llm-stats.com/
benchmark_evaluation
true
source model id
kimi-k2-instruct
Original result ↗

LLM Stats · GitHub archive

0.047

source id
llmStatsArchive
leaderboard name
Humanity's Last Exam · Normalized (reported)
dimension
general
metric
Humanity's Last Exam · Normalized (reported)
rank
1
aggregation scope
status
archived
license
CC-BY-4.0
direction
higherIsBetter
conditions
language
en
benchmark
name
Humanity's Last Exam
language
en
modality
multimodal
verified
false
max_score
1
categories
  • general
created_at
2025-07-19T19:56:12.507693+00:00
paper_link
https://arxiv.org/abs/2501.14249
updated_at
2025-07-19T19:56:12.507693+00:00
description
A multi-modal benchmark at the frontier of human knowledge with 2,500 questions across dozens of subjects including mathematics, humanities, and natural sciences, created by nearly 1000 subject expert contributors from over 500 institutions
benchmark_id
humanity's-last-exam
multilingual
false
analysis_method
Accuracy (Text Only)
is_self_reported
true
verified_by_llmstats
false
attribution
Jonathan Chavez Tamales and LLM Stats contributors (copyright 2024 jc); TensorTrace structures original results and derives scoped ranks.
license_url
https://creativecommons.org/licenses/by/4.0/
metric_kind
Humanity's Last Exam · Normalized (reported)
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
source_file
data/organizations/moonshotai/models/kimi-k2-instruct/benchmarks.json
archive_kind
llmStats
evidence_url
https://moonshotai.github.io/Kimi-K2/
git_revision
872b75f63b8d853c46ed17856424029a0c47b1d2
published_at
2025-10-24T17:47:59Z
snapshot_date
2025-10-24T17:47:59Z
source_revision
6e27aa41f51fa8cb327602349240077d1c7acae9fead8ba2f07db0d549927ddc
publisher_anchor
moonshotai/kimi-k2-instruct
metric_population
1
current_catalog_url
https://llm-stats.com/
benchmark_evaluation
true
source model id
kimi-k2-instruct
Original result ↗

Published prices

1–6 / 6

moonshotai/kimi-k2 ↗

$0.57

Inference provider
openRouter
Platform
OpenRouter
Unit
input_tokens_per_million
Currency
USD
Billing conditions

moonshotai/kimi-k2 ↗

$2.30

Inference provider
openRouter
Platform
OpenRouter
Unit
output_tokens_per_million
Currency
USD
Billing conditions

moonshotai/kimi-k2 ↗

$0.57

Inference provider
vercelAIGatewayCatalog
Platform
novita
Unit
input_tokens_per_million
Currency
USD
Billing conditions
endpoint
novita | moonshotai/kimi-k2
model_type
language
quantization
fp8
billing_scope
via Vercel AI Gateway
Billing label
Input tokens / 1M
Pricing via
Vercel AI Gateway

moonshotai/kimi-k2 ↗

$2.30

Inference provider
vercelAIGatewayCatalog
Platform
novita
Unit
output_tokens_per_million
Currency
USD
Billing conditions
endpoint
novita | moonshotai/kimi-k2
model_type
language
quantization
fp8
billing_scope
via Vercel AI Gateway
Billing label
Output tokens / 1M
Pricing via
Vercel AI Gateway

moonshotai/kimi-k2 ↗

$0.57

Inference provider
vercelAIGatewayCatalog
Platform
Vercel AI Gateway
Unit
input_tokens_per_million
Currency
USD
Billing conditions
model_type
language
quote_scope
gateway_catalog
billing_scope
Vercel public catalog; endpoint prices may differ
Billing label
Input tokens / 1M
Pricing via
Vercel AI Gateway

moonshotai/kimi-k2 ↗

$2.30

Inference provider
vercelAIGatewayCatalog
Platform
Vercel AI Gateway
Unit
output_tokens_per_million
Currency
USD
Billing conditions
model_type
language
quote_scope
gateway_catalog
billing_scope
Vercel public catalog; endpoint prices may differ
Billing label
Output tokens / 1M
Pricing via
Vercel AI Gateway

Provider specifications & access

1–4 / 4

Kimi K2 Instruct ↗

source model id
moonshotai/kimi-k2
hosting
vercel-gateway
platforms
  • Vercel AI Gateway
context length
131072
max output tokens
131072
architecture
input_modalities
  • text
output_modalities
  • text
capabilities
  • toolCalling

Kimi-K2-Instruct ↗

source model id
moonshotai/Kimi-K2-Instruct
hosting
siliconflow-hosted
platforms
  • SiliconFlow
max output tokens
131072
architecture
input_modalities
  • text
output_modalities
  • text
capabilities
  • structuredOutputs
  • toolCalling

moonshotai/Kimi-K2-Instruct ↗

source model id
publishers/hf-moonshotai/models/kimi-k2-instruct@001
hosting
vertex-self-deployed
platforms
  • Google Cloud Vertex AI
architecture
input_modalities
  • text
output_modalities
  • text
capabilities

    Kimi-K2-Instruct ↗

    source model id
    publishers/moonshotai/models/kimi-k2@kimi-k2-instruct
    hosting
    vertex-self-deployed
    platforms
    • Google Cloud Vertex AI
    architecture
    input_modalities
    • text
    output_modalities
    • text
    capabilities

      Source specification observations

      1–7 / 7

      openRouter ↗

      field
      context_length
      value
      131072
      normalized value
      131072
      scope
      source-model
      source id
      openRouter
      confidence
      1

      siliconFlow ↗

      field
      context_length
      value
      131K
      scope
      source-model
      source id
      siliconFlow
      confidence
      1

      vercelAIGatewayCatalog ↗

      field
      context_length
      value
      131072
      normalized value
      131072
      scope
      source-model
      source id
      vercelAIGatewayCatalog
      confidence
      1

      openRouter ↗

      field
      knowledge_cutoff
      reported field
      knowledge_cutoff
      value
      2024-12-31
      normalized value
      2024-12-31
      precision
      day
      scope
      source-model
      source id
      openRouter
      confidence
      1

      openRouter ↗

      field
      max_output_tokens
      value
      98304
      normalized value
      98304
      scope
      source-model
      source id
      openRouter
      confidence
      1

      siliconFlow ↗

      field
      max_output_tokens
      value
      131K
      scope
      source-model
      source id
      siliconFlow
      confidence
      1

      vercelAIGatewayCatalog ↗

      field
      max_output_tokens
      value
      131072
      normalized value
      131072
      scope
      source-model
      source id
      vercelAIGatewayCatalog
      confidence
      1

      Original sources

      1–19 / 19

      SiliconFlow · Model page ↗

      moonshotai.github.io · Model reference ↗

      Hugging Face · Model repository ↗

      Google Cloud · Model page ↗

      Google Cloud · Model page ↗

      OpenRouter · Model page ↗

      SiliconFlow · Playground ↗

      Vercel AI Gateway · Model page ↗

      SiliconFlow · API documentation ↗

      Epoch AI · Catalog ↗

      GitHub · Catalog ↗

      Google Cloud · Catalog ↗

      Hugging Face · Catalog ↗

      Hugging Face · Catalog ↗

      OpenRouter · Catalog ↗

      SiliconFlow · Catalog ↗

      Vercel AI Gateway · Catalog ↗

      crfm.stanford.edu · Catalog ↗

      www.swebench.com · Catalog ↗

      Source links

      1–12 / 12

      Explore more models

      Models

      Loading
      EVIDENCE / COMPARISON

      Benchmarks

      Explore every benchmark

      Loading

      These charts contain model-linked results. The charts above also include unlinked source profiles. Compare scores within each evaluation. Model quality, usage, transparency and system performance measure different things.

      Install TensorTrace