GPT-4o (May 2024): benchmark scores, price and ranking
GPT-4o (May 2024) from OpenAI ranks #643 of 743 LLMs on the LLMs Tiger aggregate benchmark score (44.3), based on 7 benchmark results captured up to 2026-10-13.
- Organisation
- OpenAI
- Type
- Frontier model
- Open weights
- No
- Released
- 2024-05-13 (source: epoch)
- Cheapest API price
- $2.50 in / $10.00 out per million tokens (via azure, observed 2026-10-13T20:47Z)
- Aggregate rank
- #643 of 743 (score 44.3)
Relative to other models, GPT-4o (May 2024) is strongest on BALROG (#12 of 34) and weakest on OTIS Mock AIME (#152 of 189).
Across the 7 benchmarks where at least ten models are scored, GPT-4o (May 2024) is in the top 10% on 0 and in the top half on 2.
Its cheapest listed input price, $2.50 per million tokens, is lower than that of 15% of the 427 models with a listed price.
GPT-4o (May 2024) benchmark results
| Benchmark | Score | Rank | Source | Captured |
|---|---|---|---|---|
| BALROG | 32.3% | #12 of 34 | Epoch AI Benchmarking Hub | 2026-10-13 |
| Epoch ECI | 129 | #94 of 204 | Epoch AI Benchmarking Hub | 2026-10-13 |
| DTBench | 64.5% | #85 of 146 | Epoch AI Benchmarking Hub | 2026-10-13 |
| METR time horizon | 9 min | #21 of 31 | Epoch AI Benchmarking Hub | 2026-10-13 |
| GPQA Diamond | 48.9% | #192 of 275 | Epoch AI Benchmarking Hub | 2026-10-13 |
| LMCA | 16.6% | #83 of 108 | Epoch AI Benchmarking Hub | 2026-10-13 |
| OTIS Mock AIME | 6.2% | #152 of 189 | Epoch AI Benchmarking Hub | 2026-10-13 |
About GPT-4o (May 2024)
GPT-4o (May 2024) is a frontier model from OpenAI. It was released on 2024-05-13, according to Epoch AI. The cheapest API listing we track charges $2.50 per million input tokens and $10.00 per million output tokens, through azure. We hold 7 benchmark results for it across 3 areas, and none is in the top tenth of its board.
Where GPT-4o (May 2024) stands
On our aggregate score, GPT-4o (May 2024) is #643 of 743 ranked models, in the lower half. Its score is 44.3; the scale is centred on 50, so it sits 5.7 points below the typical model. That beats 13% of the ranked field. Among the 88 ranked OpenAI models it is #75. The next one up is o1-mini-2024-09-12 at 44.8. The next one down is gpt-oss-20b (high) at 44.3.
GPT-4o (May 2024), benchmark by benchmark
BALROG puts GPT-4o (May 2024) at 32.3%, #12 of 34. That beats 65% of the models tested, but 36.0 points short of GPT-6 Astra (max) (68.3%). The results either side are Claude 3.5 Sonnet (Oct 2024) (32.6%) and claude-haiku-4-5-20251001 (31.2%). It covers long-horizon game-playing agents. The board median is 27.6%, so GPT-4o (May 2024) is 4.7 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
On Epoch ECI, GPT-4o (May 2024) reaches 129. It ranks #94 of 204, ahead of 54% of the field, and 27 points behind Qwen 3.8 Max at 156. The results either side are Qwen2.5-72B (129) and GPT-4o (Nov 2024) (129). It covers Epoch Capabilities Index: Epoch's own aggregate of ~40 benchmarks. The board median is 127, so GPT-4o (May 2024) is 2 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
GPT-4o (May 2024) lands below the middle on DTBench: 64.5%, #85 of 146, with 58% of the field scoring higher. Claude Opus 5.5 (max) leads at 98.9%; the closest result above is DeepSeek-V3 (Mar 2025) at 64.8%. It covers dynamic tool-use benchmark. The board median is 68.8%, so GPT-4o (May 2024) is 4.3 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
GPT-4o (May 2024) lands below the middle on METR time horizon: 9 min, #21 of 31, with 65% of the field scoring higher. Claude Mythos Preview (Early) leads at 1,045 min; the closest result above is Claude 3.5 Sonnet (Jun 2024) at 11 min. It covers length of task (minutes) done at 50% success. The board median is 23 min, so GPT-4o (May 2024) is 14 minutes below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
GPT-4o (May 2024) lands below the middle on GPQA Diamond: 48.9%, #192 of 275, with 69% of the field scoring higher. GPT-6 Astra (max) leads at 95.8%; the closest result above is GPT-4.1 nano at 48.9%. It covers expert-level science questions. The board median is 71.7%, so GPT-4o (May 2024) is 22.8 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
GPT-4o (May 2024) lands below the middle on LMCA: 16.6%, #83 of 108, with 76% of the field scoring higher. Claude Opus 5.5 (max) leads at 68.2%; the closest result above is Mistral Large 3 at 16.7%. It covers language-model capability assessment. The board median is 29.2%, so GPT-4o (May 2024) is 12.6 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
OTIS Mock AIME is a weak spot. GPT-4o (May 2024) scores 6.2%, #152 of 189, behind 80% of the models measured. The leader, Claude Fable 5.1 (max), reaches 100.0%. It covers competition maths (AIME 2024-25 style). The board median is 57.8%, so GPT-4o (May 2024) is 51.5 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Strengths and weaknesses
Reasoning, maths and science: weak, ahead of 25% of the field on average across 2 benchmarks. Its best board here is GPQA Diamond (#192 of 275) and its weakest is OTIS Mock AIME (#152 of 189). Agents and tool use: solid, ahead of 53% of the field on average across 2 benchmarks. Its best board here is BALROG (#12 of 34) and its weakest is DTBench (#85 of 146). General knowledge and chat: uneven, ahead of 36% of the field on average across 3 benchmarks. Its best board here is Epoch ECI (#94 of 204) and its weakest is LMCA (#83 of 108).
Price and value
The median input price across the 427 priced models is $0.500 per million tokens, so GPT-4o (May 2024) costs more than the typical listing. Models that score within 1.5 points or better and cost less: Qwen2.5-Coder-7B-Instruct ($0.010, 49.5), DeepSeek V4.1 Flash (high) ($0.010, 48.9), Qwen2.5-Coder-3B-Instruct ($0.010, 46.5). Similar scores cost more on Claude 3.5 Sonnet (Oct 2024) ($3.00) and Claude 3.5 Sonnet (Jun 2024) ($3.00).
Release history
The OpenAI model released before it was gpt-4o-2024-05-13 on 2024-05-13, which scores 51.9. The next one was GPT-4o mini on 2024-07-18, at 47.1.
Reading these numbers
The 7 scores on this page come from 1 source: Epoch AI Benchmarking Hub (7). No official leaderboard supplied any of them, so treat them as third-party measurements. With 7 results the aggregate rank can move a lot when one new score arrives.
Models ranked nearby
- #640 Laguna M.1 (44.4)
- #641 grok-code-fast-1 (44.3)
- #642 Qwen1.5-7B-Chat (44.3)
- #644 gpt-oss-20b (high) (44.3)
- #645 Yi-6B-Chat (44.3)
- #646 Claude 3.5 Sonnet (Jun 2024) (44.2)
More from OpenAI
- GPT-6 Astra (max) (#2, 64.0)
- GPT-6.1 Sol (max) (#5, 60.6)
- GPT-5.6 Sol (max) (#6, 60.1)
- GPT-6 Sol (max) (#7, 60.0)
- GPT-5.5 (xhigh) (#11, 58.0)
- GPT-5.6 Terra (max) (#12, 57.8)
Compare it on the live leaderboard → · How the aggregate score works