Compare · ModelsLive · 2 picked · head to head
DeepSeek V4 Flash 0731 vs Gemini 2.5 Pro
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
DeepSeek V4 Flash 0731 wins on 10/12 benchmarks
DeepSeek V4 Flash 0731 wins 10 of 12 shared benchmarks. Leads in reasoning · knowledge · general.
Category leads
reasoning·DeepSeek V4 Flash 0731knowledge·DeepSeek V4 Flash 0731general·DeepSeek V4 Flash 0731math·DeepSeek V4 Flash 0731coding·DeepSeek V4 Flash 0731
Hype vs Reality
Attention vs performance
DeepSeek V4 Flash 0731
#76 by perf·#6 by attention
Gemini 2.5 Pro
#116 by perf·no signal
Best value
DeepSeek V4 Flash 0731
9.5x better value than Gemini 2.5 Pro
DeepSeek V4 Flash 0731
85.5 pts/$
$0.65/M
Gemini 2.5 Pro
9.0 pts/$
$5.63/M
Vendor risk
Mixed exposure
One or more vendors flagged
DeepSeek
$3.4B·Tier 1
Google DeepMind
$4.20T·Tier 1
Head to head
12 benchmarks · 2 models
DeepSeek V4 Flash 0731Gemini 2.5 Pro
ARC-AGI
DeepSeek V4 Flash 0731 leads by +48.0
ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization.
DeepSeek V4 Flash 0731
89.0
Gemini 2.5 Pro
41.0
ARC-AGI-2
DeepSeek V4 Flash 0731 leads by +56.5
ARC-AGI-2 · the second iteration of the Abstraction and Reasoning Corpus, testing novel pattern recognition and abstract reasoning without prior training data.
DeepSeek V4 Flash 0731
61.4
Gemini 2.5 Pro
4.9
Chess Puzzles
DeepSeek V4 Flash 0731 leads by +13.7
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
DeepSeek V4 Flash 0731
29.5
Gemini 2.5 Pro
15.8
Dtbench
DeepSeek V4 Flash 0731 leads by +14.2
DeepSeek V4 Flash 0731
84.9
Gemini 2.5 Pro
70.7
FrontierMath-Tier-4-v2-Private
DeepSeek V4 Flash 0731 leads by +24.4
DeepSeek V4 Flash 0731
24.4
Gemini 2.5 Pro
0.0
FrontierMath-Tiers-1-3-v2-Private
DeepSeek V4 Flash 0731 leads by +33.0
DeepSeek V4 Flash 0731
57.5
Gemini 2.5 Pro
24.6
GPQA diamond
DeepSeek V4 Flash 0731 leads by +7.7
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
DeepSeek V4 Flash 0731
88.0
Gemini 2.5 Pro
80.4
Lmca
DeepSeek V4 Flash 0731 leads by +8.2
DeepSeek V4 Flash 0731
49.1
Gemini 2.5 Pro
40.9
OTIS Mock AIME 2024-2025
DeepSeek V4 Flash 0731 leads by +9.7
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
DeepSeek V4 Flash 0731
94.4
Gemini 2.5 Pro
84.7
SimpleBench
Gemini 2.5 Pro leads by +1.6
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
DeepSeek V4 Flash 0731
53.3
Gemini 2.5 Pro
54.9
SimpleQA Verified
Gemini 2.5 Pro leads by +22.4
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
DeepSeek V4 Flash 0731
33.6
Gemini 2.5 Pro
56.0
WeirdML
DeepSeek V4 Flash 0731 leads by +8.9
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
DeepSeek V4 Flash 0731
63.0
Gemini 2.5 Pro
54.0
Full benchmark table
| Benchmark | DeepSeek V4 Flash 0731 | Gemini 2.5 Pro |
|---|---|---|
ARC-AGI ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization. | 89.0 | 41.0 |
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. | 61.4 | 4.9 |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 29.5 | 15.8 |
Dtbench | 84.9 | 70.7 |
FrontierMath-Tier-4-v2-Private | 24.4 | 0.0 |
FrontierMath-Tiers-1-3-v2-Private | 57.5 | 24.6 |
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 | 80.4 |
Lmca | 49.1 | 40.9 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 94.4 | 84.7 |
SimpleBench SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking. | 53.3 | 54.9 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 33.6 | 56.0 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 63.0 | 54.0 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $0.02 | $1.28 | 1.0M tokens (~524 books) | $3.31 | |
| $1.25 | $10.00 | 1.0M tokens (~524 books) | $34.38 |