Compare · ModelsLive · 2 picked · head to head
DeepSeek V4 Pro 0813 vs Gemini 3.7 Flash
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Gemini 3.7 Flash wins on 11/14 benchmarks
Gemini 3.7 Flash wins 11 of 14 shared benchmarks. Leads in agentic · reasoning · knowledge.
Category leads
agentic·Gemini 3.7 Flashreasoning·Gemini 3.7 Flashknowledge·Gemini 3.7 Flashgeneral·Gemini 3.7 Flashcoding·Gemini 3.7 Flashmath·Gemini 3.7 Flash
Hype vs Reality
Attention vs performance
DeepSeek V4 Pro 0813
#47 by perf·#6 by attention
Gemini 3.7 Flash
#60 by perf·#5 by attention
Best value
Gemini 3.7 Flash
1.2x better value than DeepSeek V4 Pro 0813
DeepSeek V4 Pro 0813
21.6 pts/$
$2.77/M
Gemini 3.7 Flash
25.6 pts/$
$2.25/M
Vendor risk
Mixed exposure
One or more vendors flagged
DeepSeek
$3.4B·Tier 1
Google DeepMind
$4.20T·Tier 1
Head to head
14 benchmarks · 2 models
DeepSeek V4 Pro 0813Gemini 3.7 Flash
APEX-Agents
Gemini 3.7 Flash leads by +20.5
APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments.
DeepSeek V4 Pro 0813
47.3
Gemini 3.7 Flash
67.8
ARC-AGI
Gemini 3.7 Flash leads by +5.0
ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization.
DeepSeek V4 Pro 0813
90.5
Gemini 3.7 Flash
95.5
ARC-AGI-2
Gemini 3.7 Flash leads by +23.3
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 Pro 0813
61.3
Gemini 3.7 Flash
84.6
Chess Puzzles
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
DeepSeek V4 Pro 0813
44.2
Gemini 3.7 Flash
44.2
Dtbench
Gemini 3.7 Flash leads by +4.9
DeepSeek V4 Pro 0813
89.8
Gemini 3.7 Flash
94.7
Frontiercode
Gemini 3.7 Flash leads by +15.0
DeepSeek V4 Pro 0813
28.6
Gemini 3.7 Flash
43.6
FrontierMath-Tier-4-v2-Private
Gemini 3.7 Flash leads by +9.8
DeepSeek V4 Pro 0813
26.8
Gemini 3.7 Flash
36.6
FrontierMath-Tiers-1-3-v2-Private
Gemini 3.7 Flash leads by +7.0
DeepSeek V4 Pro 0813
64.6
Gemini 3.7 Flash
71.6
GPQA diamond
Gemini 3.7 Flash leads by +4.2
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
DeepSeek V4 Pro 0813
88.9
Gemini 3.7 Flash
93.1
Lmca
Gemini 3.7 Flash leads by +5.8
DeepSeek V4 Pro 0813
53.5
Gemini 3.7 Flash
59.3
Mystery Game Puzzles
DeepSeek V4 Pro 0813 leads by +6.6
DeepSeek V4 Pro 0813
37.2
Gemini 3.7 Flash
30.6
OTIS Mock AIME 2024-2025
DeepSeek V4 Pro 0813 leads by +1.4
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
DeepSeek V4 Pro 0813
98.6
Gemini 3.7 Flash
97.2
Proofbench
Gemini 3.7 Flash leads by +8.0
DeepSeek V4 Pro 0813
50.0
Gemini 3.7 Flash
58.0
SimpleQA Verified
Gemini 3.7 Flash leads by +16.3
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
DeepSeek V4 Pro 0813
52.9
Gemini 3.7 Flash
69.2
Full benchmark table
| Benchmark | DeepSeek V4 Pro 0813 | Gemini 3.7 Flash |
|---|---|---|
APEX-Agents APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments. | 47.3 | 67.8 |
ARC-AGI ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization. | 90.5 | 95.5 |
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. | 61.3 | 84.6 |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 44.2 | 44.2 |
Dtbench | 89.8 | 94.7 |
Frontiercode | 28.6 | 43.6 |
FrontierMath-Tier-4-v2-Private | 26.8 | 36.6 |
FrontierMath-Tiers-1-3-v2-Private | 64.6 | 71.6 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 88.9 | 93.1 |
Lmca | 53.5 | 59.3 |
Mystery Game Puzzles | 37.2 | 30.6 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 98.6 | 97.2 |
Proofbench | 50.0 | 58.0 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 52.9 | 69.2 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $0.55 | $5.00 | 1.0M tokens (~524 books) | $16.63 | |
| $0.75 | $3.75 | 1.0M tokens (~524 books) | $15.00 |
People also compared
Gemini 3.7 Flash vs GPT-5.5 ProGemini 3.7 Flash vs GPT-5.5Claude Opus 5.5 vs Gemini 3.7 FlashClaude Mythos Preview vs Gemini 3.7 FlashDeepSeek V3.2 Speciale vs DeepSeek V4 Pro 0813DeepSeek-V2 (MoE-236B, May 2024) vs DeepSeek V4 Pro 0813DeepSeek V3.2 Speciale vs Gemini 3.7 FlashDeepSeek-V2 (MoE-236B, May 2024) vs Gemini 3.7 Flash