Compare · ModelsLive · 2 picked · head to head
Claude Opus 4.8 vs Kimi K3
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Claude Opus 4.8 wins on 13/20 benchmarks
Claude Opus 4.8 wins 13 of 20 shared benchmarks. Leads in reasoning · coding · general.
Category leads
agentic·Kimi K3reasoning·Claude Opus 4.8knowledge·Kimi K3coding·Claude Opus 4.8general·Claude Opus 4.8math·Claude Opus 4.8
Hype vs Reality
Attention vs performance
Claude Opus 4.8
#75 by perf·#9 by attention
Kimi K3
#51 by perf·#17 by attention
Best value
Kimi K3
2.2x better value than Claude Opus 4.8
Claude Opus 4.8
3.7 pts/$
$15.00/M
Kimi K3
8.1 pts/$
$7.33/M
Vendor risk
Who is behind the model
Anthropic
$965.0B·Tier 1
Moonshot AI
$18.0B·Tier 1
Head to head
20 benchmarks · 2 models
Claude Opus 4.8Kimi K3
APEX-Agents
Kimi K3 leads by +1.7
APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments.
Claude Opus 4.8
48.9
Kimi K3
50.6
ARC-AGI
Kimi K3 leads by +2.0
ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization.
Claude Opus 4.8
92.5
Kimi K3
94.5
ARC-AGI-2
Claude Opus 4.8 leads by +11.7
ARC-AGI-2 · the second iteration of the Abstraction and Reasoning Corpus, testing novel pattern recognition and abstract reasoning without prior training data.
Claude Opus 4.8
72.1
Kimi K3
60.4
Chess Puzzles
Kimi K3 leads by +5.3
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
Claude Opus 4.8
30.6
Kimi K3
35.8
Deepswe
Kimi K3 leads by +9.5
Claude Opus 4.8
59.0
Kimi K3
68.5
Dtbench
Claude Opus 4.8 leads by +6.2
Claude Opus 4.8
91.5
Kimi K3
85.3
Frontiercode
Claude Opus 4.8 leads by +2.3
Claude Opus 4.8
46.5
Kimi K3
44.2
FrontierMath-Tier-4-v2-Private
Claude Opus 4.8 leads by +17.1
Claude Opus 4.8
56.1
Kimi K3
39.0
FrontierMath-Tiers-1-3-v2-Private
Claude Opus 4.8 leads by +7.8
Claude Opus 4.8
80.0
Kimi K3
72.2
Furniture Assembly
Claude Opus 4.8 leads by +11.9
Claude Opus 4.8
17.9
Kimi K3
6.0
GPQA diamond
Kimi K3 leads by +2.8
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Claude Opus 4.8
88.0
Kimi K3
90.8
Lmca
Claude Opus 4.8 leads by +5.6
Claude Opus 4.8
67.7
Kimi K3
62.0
Mystery Game Puzzles
Claude Opus 4.8 leads by +11.0
Claude Opus 4.8
29.5
Kimi K3
18.5
OTIS Mock AIME 2024-2025
Claude Opus 4.8 leads by +1.1
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Claude Opus 4.8
98.3
Kimi K3
97.2
PostTrainBench
Claude Opus 4.8 leads by +1.9
Claude Opus 4.8
33.8
Kimi K3
32.0
Proofbench
Kimi K3 leads by +18.0
Claude Opus 4.8
69.0
Kimi K3
87.0
SimpleBench
Claude Opus 4.8 leads by +4.9
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
Claude Opus 4.8
57.8
Kimi K3
52.8
SimpleQA Verified
Claude Opus 4.8 leads by +2.4
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
Claude Opus 4.8
53.0
Kimi K3
50.6
Surface Evolver Bench
Kimi K3 leads by +7.5
Claude Opus 4.8
87.5
Kimi K3
95.0
WeirdML
Claude Opus 4.8 leads by +0.3
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
Claude Opus 4.8
82.9
Kimi K3
82.6
Full benchmark table
| Benchmark | Claude Opus 4.8 | Kimi K3 |
|---|---|---|
APEX-Agents APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments. | 48.9 | 50.6 |
ARC-AGI ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization. | 92.5 | 94.5 |
ARC-AGI-2 ARC-AGI-2 · the second iteration of the Abstraction and Reasoning Corpus, testing novel pattern recognition and abstract reasoning without prior training data. | 72.1 | 60.4 |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 30.6 | 35.8 |
Deepswe | 59.0 | 68.5 |
Dtbench | 91.5 | 85.3 |
Frontiercode | 46.5 | 44.2 |
FrontierMath-Tier-4-v2-Private | 56.1 | 39.0 |
FrontierMath-Tiers-1-3-v2-Private | 80.0 | 72.2 |
Furniture Assembly | 17.9 | 6.0 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 88.0 | 90.8 |
Lmca | 67.7 | 62.0 |
Mystery Game Puzzles | 29.5 | 18.5 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 98.3 | 97.2 |
PostTrainBench | 33.8 | 32.0 |
Proofbench | 69.0 | 87.0 |
SimpleBench SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking. | 57.8 | 52.8 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 53.0 | 50.6 |
Surface Evolver Bench | 87.5 | 95.0 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 82.9 | 82.6 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $5.00 | $25.00 | 1.0M tokens (~500 books) | $100.00 | |
| $0.67 | $14.00 | 1.0M tokens (~524 books) | $40.03 |