Compare · ModelsLive · 2 picked · head to head
GLM 4.7 vs Kimi K2 Thinking
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Kimi K2 Thinking wins on 13/23 benchmarks
Kimi K2 Thinking wins 13 of 23 shared benchmarks. Leads in agentic · knowledge · language.
Category leads
agentic·Kimi K2 Thinkingknowledge·Kimi K2 Thinkingmath·GLM 4.7coding·GLM 4.7reasoning·GLM 4.7language·Kimi K2 Thinking
Hype vs Reality
Attention vs performance
GLM 4.7
#91 by perf·no signal
Kimi K2 Thinking
#77 by perf·no signal
Best value
GLM 4.7
1.4x better value than Kimi K2 Thinking
GLM 4.7
47.2 pts/$
$1.07/M
Kimi K2 Thinking
34.4 pts/$
$1.55/M
Vendor risk
Who is behind the model
z-ai
private · undisclosed
moonshotai
private · undisclosed
Head to head
23 benchmarks · 2 models
GLM 4.7Kimi K2 Thinking
APEX-Agents
Kimi K2 Thinking leads by +0.9
APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments.
GLM 4.7
3.1
Kimi K2 Thinking
4.0
Chess Puzzles
Kimi K2 Thinking leads by +14.0
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
GLM 4.7
6.0
Kimi K2 Thinking
20.0
FrontierMath-2025-02-28-Private
Kimi K2 Thinking leads by +19.0
FrontierMath (Feb 2025) · original research-level math problems created by mathematicians, testing capabilities at the boundary of current AI mathematical reasoning.
GLM 4.7
2.4
Kimi K2 Thinking
21.4
FrontierMath-Tier-4-2025-07-01-Private
FrontierMath Tier 4 (Jul 2025) · the most challenging tier of frontier mathematics, containing problems that push the absolute limits of AI mathematical reasoning.
GLM 4.7
0.1
Kimi K2 Thinking
0.1
GPQA diamond
Kimi K2 Thinking leads by +1.2
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
GLM 4.7
77.8
Kimi K2 Thinking
79.0
LiveBench · Agentic Coding
GLM 4.7 leads by +3.3
GLM 4.7
41.7
Kimi K2 Thinking
38.3
LiveBench · Coding
GLM 4.7 leads by +5.7
GLM 4.7
73.1
Kimi K2 Thinking
67.4
LiveBench · Data Analysis
GLM 4.7 leads by +2.9
GLM 4.7
55.2
Kimi K2 Thinking
52.3
LiveBench · If
Kimi K2 Thinking leads by +26.4
GLM 4.7
35.7
Kimi K2 Thinking
62.0
LiveBench · Language
Kimi K2 Thinking leads by +1.2
GLM 4.7
65.2
Kimi K2 Thinking
66.5
LiveBench · Mathematics
Kimi K2 Thinking leads by +5.1
GLM 4.7
76.0
Kimi K2 Thinking
81.1
LiveBench · Overall
Kimi K2 Thinking leads by +3.5
GLM 4.7
58.1
Kimi K2 Thinking
61.6
LiveBench · Reasoning
Kimi K2 Thinking leads by +3.8
GLM 4.7
59.7
Kimi K2 Thinking
63.5
OpenCompass · AIME2025
GLM 4.7 leads by +1.3
GLM 4.7
95.4
Kimi K2 Thinking
94.1
OpenCompass · GPQA-Diamond
GLM 4.7 leads by +4.2
GLM 4.7
86.9
Kimi K2 Thinking
82.7
OpenCompass · HLE
GLM 4.7 leads by +4.1
GLM 4.7
25.4
Kimi K2 Thinking
21.3
OpenCompass · IFEval
Kimi K2 Thinking leads by +2.2
GLM 4.7
90.2
Kimi K2 Thinking
92.4
OpenCompass · LiveCodeBenchV6
GLM 4.7 leads by +6.7
GLM 4.7
83.8
Kimi K2 Thinking
77.1
OpenCompass · MMLU-Pro
Kimi K2 Thinking leads by +0.3
GLM 4.7
84.0
Kimi K2 Thinking
84.3
OTIS Mock AIME 2024-2025
GLM 4.7 leads by +0.3
OTIS Mock AIME 2024–2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
GLM 4.7
83.3
Kimi K2 Thinking
83.0
PostTrainBench
GLM 4.7 leads by +0.2
GLM 4.7
7.5
Kimi K2 Thinking
7.3
SimpleQA Verified
Kimi K2 Thinking leads by +0.1
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
GLM 4.7
31.5
Kimi K2 Thinking
31.6
Terminal Bench
Kimi K2 Thinking leads by +2.3
Terminal Bench · tests the ability to accomplish real-world tasks using terminal commands, evaluating shell scripting and CLI tool proficiency.
GLM 4.7
33.4
Kimi K2 Thinking
35.7
Full benchmark table
| Benchmark | GLM 4.7 | Kimi K2 Thinking |
|---|---|---|
APEX-Agents APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments. | 3.1 | 4.0 |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 6.0 | 20.0 |
FrontierMath-2025-02-28-Private FrontierMath (Feb 2025) · original research-level math problems created by mathematicians, testing capabilities at the boundary of current AI mathematical reasoning. | 2.4 | 21.4 |
FrontierMath-Tier-4-2025-07-01-Private FrontierMath Tier 4 (Jul 2025) · the most challenging tier of frontier mathematics, containing problems that push the absolute limits of AI mathematical reasoning. | 0.1 | 0.1 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 77.8 | 79.0 |
LiveBench · Agentic Coding | 41.7 | 38.3 |
LiveBench · Coding | 73.1 | 67.4 |
LiveBench · Data Analysis | 55.2 | 52.3 |
LiveBench · If | 35.7 | 62.0 |
LiveBench · Language | 65.2 | 66.5 |
LiveBench · Mathematics | 76.0 | 81.1 |
LiveBench · Overall | 58.1 | 61.6 |
LiveBench · Reasoning | 59.7 | 63.5 |
OpenCompass · AIME2025 | 95.4 | 94.1 |
OpenCompass · GPQA-Diamond | 86.9 | 82.7 |
OpenCompass · HLE | 25.4 | 21.3 |
OpenCompass · IFEval | 90.2 | 92.4 |
OpenCompass · LiveCodeBenchV6 | 83.8 | 77.1 |
OpenCompass · MMLU-Pro | 84.0 | 84.3 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024–2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 83.3 | 83.0 |
PostTrainBench | 7.5 | 7.3 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 31.5 | 31.6 |
Terminal Bench Terminal Bench · tests the ability to accomplish real-world tasks using terminal commands, evaluating shell scripting and CLI tool proficiency. | 33.4 | 35.7 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| GLM 4.7 | $0.39 | $1.75 | 203K tokens (~101 books) | $7.30 |
| Kimi K2 Thinking | $0.60 | $2.50 | 262K tokens (~131 books) | $10.75 |
People also compared
GPT-5 Chat vs Kimi K2 ThinkingClaude Mythos Preview vs Kimi K2 ThinkingKimi K2 Thinking vs Qwen3.5 397B A17BDeepSeek V3.2 Speciale vs Kimi K2 ThinkingClaude Instant vs Kimi K2 ThinkingDeepSeek-V2 (MoE-236B, May 2024) vs Kimi K2 ThinkingGPT-5.1-Codex-Max vs Kimi K2 ThinkingKimi K2 Thinking vs Qwen3.6 Plus