Compare · ModelsLive · 2 picked · head to head
DeepSeek V3 vs Kimi K2.5
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Kimi K2.5 wins on 6/6 benchmarks
Kimi K2.5 wins 6 of 6 shared benchmarks. Leads in knowledge · math · reasoning.
Category leads
knowledge·Kimi K2.5math·Kimi K2.5reasoning·Kimi K2.5coding·Kimi K2.5
Hype vs Reality
Attention vs performance
DeepSeek V3
#45 by perf·no signal
Kimi K2.5
#87 by perf·no signal
Best value
DeepSeek V3
2.3x better value than Kimi K2.5
DeepSeek V3
97.5 pts/$
$0.60/M
Kimi K2.5
42.6 pts/$
$1.22/M
Vendor risk
Mixed exposure
One or more vendors flagged
DeepSeek
$3.4B·Tier 1
moonshotai
private · undisclosed
Head to head
6 benchmarks · 2 models
DeepSeek V3Kimi K2.5
Fiction.LiveBench
Kimi K2.5 leads by +36.1
Fiction.LiveBench · a continuously updated benchmark using recently published fiction to test reading comprehension and reasoning, preventing data contamination.
DeepSeek V3
50.0
Kimi K2.5
86.1
FrontierMath-2025-02-28-Private
Kimi K2.5 leads by +26.2
FrontierMath (Feb 2025) · original research-level math problems created by mathematicians, testing capabilities at the boundary of current AI mathematical reasoning.
DeepSeek V3
1.7
Kimi K2.5
27.9
GPQA diamond
Kimi K2.5 leads by +41.4
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
DeepSeek V3
42.0
Kimi K2.5
83.5
OTIS Mock AIME 2024-2025
Kimi K2.5 leads by +76.4
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
DeepSeek V3
15.8
Kimi K2.5
92.2
SimpleBench
Kimi K2.5 leads by +33.5
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
DeepSeek V3
2.7
Kimi K2.5
36.2
WeirdML
Kimi K2.5 leads by +9.5
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
DeepSeek V3
36.1
Kimi K2.5
45.6
Full benchmark table
| Benchmark | DeepSeek V3 | Kimi K2.5 |
|---|---|---|
Fiction.LiveBench Fiction.LiveBench · a continuously updated benchmark using recently published fiction to test reading comprehension and reasoning, preventing data contamination. | 50.0 | 86.1 |
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. | 1.7 | 27.9 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 42.0 | 83.5 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 15.8 | 92.2 |
SimpleBench SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking. | 2.7 | 36.2 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 36.1 | 45.6 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $0.32 | $0.89 | 164K tokens (~82 books) | $4.63 | |
| $0.44 | $2.00 | 262K tokens (~131 books) | $8.30 |