Compare · ModelsLive · 2 picked · head to head
Kimi K2.7 Code vs Kimi K2 Thinking
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Kimi K2.7 Code wins on 12/14 benchmarks
Kimi K2.7 Code wins 12 of 14 shared benchmarks. Leads in agentic · knowledge · coding.
Category leads
agentic·Kimi K2.7 Codeknowledge·Kimi K2.7 Codecoding·Kimi K2.7 Codereasoning·Kimi K2.7 Codelanguage·Kimi K2 Thinkingmath·Kimi K2 Thinking
Hype vs Reality
Attention vs performance
Kimi K2.7 Code
#69 by perf·#17 by attention
Kimi K2 Thinking
#103 by perf·#17 by attention
Best value
Kimi K2 Thinking
1.1x better value than Kimi K2.7 Code
Kimi K2.7 Code
30.5 pts/$
$1.84/M
Kimi K2 Thinking
34.0 pts/$
$1.55/M
Vendor risk
Who is behind the model
Moonshot AI
$18.0B·Tier 1
Moonshot AI
$18.0B·Tier 1
Head to head
14 benchmarks · 2 models
Kimi K2.7 CodeKimi K2 Thinking
APEX-Agents
Kimi K2.7 Code leads by +33.6
APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments.
Kimi K2.7 Code
37.6
Kimi K2 Thinking
4.0
Chess Puzzles
Kimi K2.7 Code leads by +1.1
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
Kimi K2.7 Code
16.9
Kimi K2 Thinking
15.8
GPQA diamond
Kimi K2.7 Code leads by +4.9
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Kimi K2.7 Code
83.8
Kimi K2 Thinking
79.0
LiveBench · Agentic Coding
Kimi K2.7 Code leads by +31.7
Kimi K2.7 Code
70.0
Kimi K2 Thinking
38.3
LiveBench · Coding
Kimi K2.7 Code leads by +6.5
Kimi K2.7 Code
74.0
Kimi K2 Thinking
67.4
LiveBench · Data Analysis
Kimi K2.7 Code leads by +10.4
Kimi K2.7 Code
62.7
Kimi K2 Thinking
52.3
LiveBench · If
Kimi K2 Thinking leads by +5.7
Kimi K2.7 Code
56.3
Kimi K2 Thinking
62.0
LiveBench · Language
Kimi K2.7 Code leads by +11.5
Kimi K2.7 Code
77.9
Kimi K2 Thinking
66.5
LiveBench · Mathematics
Kimi K2 Thinking leads by +1.5
Kimi K2.7 Code
79.6
Kimi K2 Thinking
81.1
LiveBench · Overall
Kimi K2.7 Code leads by +10.3
Kimi K2.7 Code
71.9
Kimi K2 Thinking
61.6
LiveBench · Reasoning
Kimi K2.7 Code leads by +19.3
Kimi K2.7 Code
82.8
Kimi K2 Thinking
63.5
OTIS Mock AIME 2024-2025
Kimi K2.7 Code leads by +12.5
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Kimi K2.7 Code
95.5
Kimi K2 Thinking
83.0
SimpleQA Verified
Kimi K2.7 Code leads by +4.9
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
Kimi K2.7 Code
36.5
Kimi K2 Thinking
31.6
WeirdML
Kimi K2.7 Code leads by +11.3
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
Kimi K2.7 Code
54.1
Kimi K2 Thinking
42.8
Full benchmark table
| Benchmark | Kimi K2.7 Code | 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. | 37.6 | 4.0 |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 16.9 | 15.8 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 83.8 | 79.0 |
LiveBench · Agentic Coding | 70.0 | 38.3 |
LiveBench · Coding | 74.0 | 67.4 |
LiveBench · Data Analysis | 62.7 | 52.3 |
LiveBench · If | 56.3 | 62.0 |
LiveBench · Language | 77.9 | 66.5 |
LiveBench · Mathematics | 79.6 | 81.1 |
LiveBench · Overall | 71.9 | 61.6 |
LiveBench · Reasoning | 82.8 | 63.5 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 95.5 | 83.0 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 36.5 | 31.6 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 54.1 | 42.8 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $0.61 | $3.07 | 262K tokens (~131 books) | $12.26 | |
| $0.60 | $2.50 | 262K tokens (~131 books) | $10.75 |