Compare · ModelsLive · 2 picked · head to head
GPT-4o-mini (2024-07-18) vs Gemini 1.5 Flash (May 2024)
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Gemini 1.5 Flash (May 2024) wins on 9/16 benchmarks
Gemini 1.5 Flash (May 2024) wins 9 of 16 shared benchmarks. Leads in knowledge · language · reasoning.
Category leads
arena·GPT-4o-mini (2024-07-18)knowledge·Gemini 1.5 Flash (May 2024)math·GPT-4o-mini (2024-07-18)language·Gemini 1.5 Flash (May 2024)reasoning·Gemini 1.5 Flash (May 2024)multimodal·Gemini 1.5 Flash (May 2024)coding·Gemini 1.5 Flash (May 2024)
Hype vs Reality
Attention vs performance
GPT-4o-mini (2024-07-18)
#125 by perf·no signal
Gemini 1.5 Flash (May 2024)
#105 by perf·no signal
Best value
GPT-4o-mini (2024-07-18)
GPT-4o-mini (2024-07-18)
115.2 pts/$
$0.38/M
Gemini 1.5 Flash (May 2024)
—
no price
Vendor risk
Who is behind the model
OpenAI
$840.0B·Tier 1
Google DeepMind
$4.00T·Tier 1
Head to head
16 benchmarks · 2 models
GPT-4o-mini (2024-07-18)Gemini 1.5 Flash (May 2024)
Chatbot Arena Elo · Overall
GPT-4o-mini (2024-07-18) leads by +32.1
GPT-4o-mini (2024-07-18)
1317.2
Gemini 1.5 Flash (May 2024)
1285.1
Balrog
GPT-4o-mini (2024-07-18) leads by +2.8
Balrog · benchmarks AI agents on text-based adventure games, testing language understanding, strategic planning, and long-horizon reasoning.
GPT-4o-mini (2024-07-18)
17.4
Gemini 1.5 Flash (May 2024)
14.6
GeoBench
Gemini 1.5 Flash (May 2024) leads by +12.0
GeoBench · tests geographic knowledge and spatial reasoning across countries, landmarks, coordinates, and geopolitical understanding.
GPT-4o-mini (2024-07-18)
64.0
Gemini 1.5 Flash (May 2024)
76.0
GPQA diamond
Gemini 1.5 Flash (May 2024) leads by +3.5
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
GPT-4o-mini (2024-07-18)
17.0
Gemini 1.5 Flash (May 2024)
20.5
GSM8K
GPT-4o-mini (2024-07-18) leads by +8.9
Grade School Math 8K · 8,500 linguistically diverse grade-school math word problems that require multi-step reasoning to solve.
GPT-4o-mini (2024-07-18)
91.3
Gemini 1.5 Flash (May 2024)
82.4
HELM · GPQA
Gemini 1.5 Flash (May 2024) leads by +6.9
GPT-4o-mini (2024-07-18)
36.8
Gemini 1.5 Flash (May 2024)
43.7
HELM · IFEval
Gemini 1.5 Flash (May 2024) leads by +4.9
GPT-4o-mini (2024-07-18)
78.2
Gemini 1.5 Flash (May 2024)
83.1
HELM · MMLU-Pro
Gemini 1.5 Flash (May 2024) leads by +7.5
GPT-4o-mini (2024-07-18)
60.3
Gemini 1.5 Flash (May 2024)
67.8
HELM · Omni-MATH
Gemini 1.5 Flash (May 2024) leads by +2.5
GPT-4o-mini (2024-07-18)
28.0
Gemini 1.5 Flash (May 2024)
30.5
HELM · WildBench
Gemini 1.5 Flash (May 2024) leads by +0.1
GPT-4o-mini (2024-07-18)
79.1
Gemini 1.5 Flash (May 2024)
79.2
MATH level 5
GPT-4o-mini (2024-07-18) leads by +27.5
MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics.
GPT-4o-mini (2024-07-18)
52.6
Gemini 1.5 Flash (May 2024)
25.1
MMLU
GPT-4o-mini (2024-07-18) leads by +5.2
Massive Multitask Language Understanding · 57 subjects spanning STEM, humanities, social sciences, and more. The standard benchmark for broad knowledge.
GPT-4o-mini (2024-07-18)
75.7
Gemini 1.5 Flash (May 2024)
70.5
OTIS Mock AIME 2024-2025
GPT-4o-mini (2024-07-18) leads by +3.1
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
GPT-4o-mini (2024-07-18)
6.8
Gemini 1.5 Flash (May 2024)
3.8
PIQA
GPT-4o-mini (2024-07-18) leads by +2.4
PIQA (Physical Interaction QA) · tests intuitive physical reasoning by asking models to select the correct approach for everyday physical tasks.
GPT-4o-mini (2024-07-18)
77.4
Gemini 1.5 Flash (May 2024)
75.0
VideoMME
Gemini 1.5 Flash (May 2024) leads by +7.3
VideoMME · multimodal benchmark testing video understanding across diverse domains, requiring temporal reasoning and cross-frame comprehension.
GPT-4o-mini (2024-07-18)
53.1
Gemini 1.5 Flash (May 2024)
60.4
WeirdML
Gemini 1.5 Flash (May 2024) leads by +13.1
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
GPT-4o-mini (2024-07-18)
11.8
Gemini 1.5 Flash (May 2024)
24.9
Full benchmark table
| Benchmark | GPT-4o-mini (2024-07-18) | Gemini 1.5 Flash (May 2024) |
|---|---|---|
Chatbot Arena Elo · Overall | 1317.2 | 1285.1 |
Balrog Balrog · benchmarks AI agents on text-based adventure games, testing language understanding, strategic planning, and long-horizon reasoning. | 17.4 | 14.6 |
GeoBench GeoBench · tests geographic knowledge and spatial reasoning across countries, landmarks, coordinates, and geopolitical understanding. | 64.0 | 76.0 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 17.0 | 20.5 |
GSM8K Grade School Math 8K · 8,500 linguistically diverse grade-school math word problems that require multi-step reasoning to solve. | 91.3 | 82.4 |
HELM · GPQA | 36.8 | 43.7 |
HELM · IFEval | 78.2 | 83.1 |
HELM · MMLU-Pro | 60.3 | 67.8 |
HELM · Omni-MATH | 28.0 | 30.5 |
HELM · WildBench | 79.1 | 79.2 |
MATH level 5 MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics. | 52.6 | 25.1 |
MMLU Massive Multitask Language Understanding · 57 subjects spanning STEM, humanities, social sciences, and more. The standard benchmark for broad knowledge. | 75.7 | 70.5 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 6.8 | 3.8 |
PIQA PIQA (Physical Interaction QA) · tests intuitive physical reasoning by asking models to select the correct approach for everyday physical tasks. | 77.4 | 75.0 |
VideoMME VideoMME · multimodal benchmark testing video understanding across diverse domains, requiring temporal reasoning and cross-frame comprehension. | 53.1 | 60.4 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 11.8 | 24.9 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $0.15 | $0.60 | 128K tokens (~64 books) | $2.62 | |
| — | — | — | — |