Compare · ModelsLive · 2 picked · head to head
Claude Sonnet 5 vs DeepSeek V4 Flash 0731
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Claude Sonnet 5 wins on 11/13 benchmarks
Claude Sonnet 5 wins 11 of 13 shared benchmarks. Leads in knowledge · general · coding.
Category leads
knowledge·Claude Sonnet 5general·Claude Sonnet 5coding·Claude Sonnet 5math·Claude Sonnet 5reasoning·DeepSeek V4 Flash 0731
Hype vs Reality
Attention vs performance
Claude Sonnet 5
#66 by perf·#10 by attention
DeepSeek V4 Flash 0731
#76 by perf·#6 by attention
Best value
DeepSeek V4 Flash 0731
9.1x better value than Claude Sonnet 5
Claude Sonnet 5
9.5 pts/$
$6.00/M
DeepSeek V4 Flash 0731
85.5 pts/$
$0.65/M
Vendor risk
Mixed exposure
One or more vendors flagged
Anthropic
$965.0B·Tier 1
DeepSeek
$3.4B·Tier 1
Head to head
13 benchmarks · 2 models
Claude Sonnet 5DeepSeek V4 Flash 0731
Chess Puzzles
Claude Sonnet 5 leads by +2.1
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
Claude Sonnet 5
31.6
DeepSeek V4 Flash 0731
29.5
Dtbench
Claude Sonnet 5 leads by +2.6
Claude Sonnet 5
87.5
DeepSeek V4 Flash 0731
84.9
Frontiercode
Claude Sonnet 5 leads by +23.9
Claude Sonnet 5
42.7
DeepSeek V4 Flash 0731
18.8
FrontierMath-Tier-4-v2-Private
Claude Sonnet 5 leads by +4.9
Claude Sonnet 5
29.3
DeepSeek V4 Flash 0731
24.4
FrontierMath-Tiers-1-3-v2-Private
Claude Sonnet 5 leads by +8.1
Claude Sonnet 5
65.6
DeepSeek V4 Flash 0731
57.5
GPQA diamond
DeepSeek V4 Flash 0731 leads by +0.7
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Claude Sonnet 5
87.4
DeepSeek V4 Flash 0731
88.0
Lmca
Claude Sonnet 5 leads by +9.8
Claude Sonnet 5
58.9
DeepSeek V4 Flash 0731
49.1
Mystery Game Puzzles
Claude Sonnet 5 leads by +1.1
Claude Sonnet 5
28.4
DeepSeek V4 Flash 0731
27.3
OTIS Mock AIME 2024-2025
Claude Sonnet 5 leads by +0.3
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Claude Sonnet 5
94.7
DeepSeek V4 Flash 0731
94.4
Proofbench
Claude Sonnet 5 leads by +21.0
Claude Sonnet 5
77.0
DeepSeek V4 Flash 0731
56.0
SimpleBench
DeepSeek V4 Flash 0731 leads by +0.6
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
Claude Sonnet 5
52.7
DeepSeek V4 Flash 0731
53.3
SimpleQA Verified
Claude Sonnet 5 leads by +0.1
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
Claude Sonnet 5
33.7
DeepSeek V4 Flash 0731
33.6
WeirdML
Claude Sonnet 5 leads by +5.8
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
Claude Sonnet 5
68.8
DeepSeek V4 Flash 0731
63.0
Full benchmark table
| Benchmark | Claude Sonnet 5 | DeepSeek V4 Flash 0731 |
|---|---|---|
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 31.6 | 29.5 |
Dtbench | 87.5 | 84.9 |
Frontiercode | 42.7 | 18.8 |
FrontierMath-Tier-4-v2-Private | 29.3 | 24.4 |
FrontierMath-Tiers-1-3-v2-Private | 65.6 | 57.5 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 87.4 | 88.0 |
Lmca | 58.9 | 49.1 |
Mystery Game Puzzles | 28.4 | 27.3 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 94.7 | 94.4 |
Proofbench | 77.0 | 56.0 |
SimpleBench SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking. | 52.7 | 53.3 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 33.7 | 33.6 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 68.8 | 63.0 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $2.00 | $10.00 | 1.0M tokens (~500 books) | $40.00 | |
| $0.02 | $1.28 | 1.0M tokens (~524 books) | $3.31 |