Gemini 2.5 Pro Exp (Mar 2025): benchmark scores, price and ranking
Gemini 2.5 Pro Exp (Mar 2025) from Google DeepMind ranks #140 of 743 LLMs on the LLMs Tiger aggregate benchmark score (51.2), based on 6 benchmark results captured up to 2026-10-13.
- Organisation
- Google DeepMind
- Type
- Preview release
- Open weights
- No
- Released
- 2025-03-25 (source: epoch)
- Aggregate rank
- #140 of 743 (score 51.2)
Relative to other models, Gemini 2.5 Pro Exp (Mar 2025) is strongest on BALROG (#8 of 34) and weakest on EnigmaEval (#9 of 24).
Across the 6 benchmarks where at least ten models are scored, Gemini 2.5 Pro Exp (Mar 2025) is in the top 10% on 0 and in the top half on 6.
Gemini 2.5 Pro Exp (Mar 2025) benchmark results
| Benchmark | Score | Rank | Source | Captured |
|---|---|---|---|---|
| BALROG | 43.3% | #8 of 34 | Epoch AI Benchmarking Hub | 2026-10-13 |
| GPQA Diamond | 83.8% | #87 of 275 | Epoch AI Benchmarking Hub | 2026-10-13 |
| Fiction.liveBench | 66.7% | #15 of 47 | Epoch AI Benchmarking Hub | 2026-10-13 |
| SimpleBench | 51.6% | #20 of 61 | Epoch AI Benchmarking Hub | 2026-10-13 |
| Humanity's Last Exam | 18.2% | #9 of 27 | Epoch AI Benchmarking Hub | 2026-10-13 |
| EnigmaEval | 4.1% | #9 of 24 | Epoch AI Benchmarking Hub | 2026-10-13 |
About Gemini 2.5 Pro Exp (Mar 2025)
Gemini 2.5 Pro Exp (Mar 2025) is a preview release from Google DeepMind. It was released on 2025-03-25, according to Epoch AI. No API price is listed for it yet. We hold 6 benchmark results for it across 3 areas, and none is in the top tenth of its board.
Where Gemini 2.5 Pro Exp (Mar 2025) stands
On our aggregate score, Gemini 2.5 Pro Exp (Mar 2025) is #140 of 743 ranked models, in the top quarter. Its score is 51.2; the scale is centred on 50, so it sits 1.2 points above the typical model. That beats 81% of the ranked field. Among the 41 ranked Google DeepMind models it is #7. The next one up is Gemini 3 Flash at 51.5. The next one down is Gemini 2.5 Pro Preview (Jun 2025) at 51.0.
Gemini 2.5 Pro Exp (Mar 2025), benchmark by benchmark
BALROG puts Gemini 2.5 Pro Exp (Mar 2025) at 43.3%, #8 of 34. That beats 76% of the models tested, but 25.0 points short of GPT-6 Astra (max) (68.3%). The results either side are grok-4-0709 (43.6%) and DeepSeek-R1 (34.9%). It covers long-horizon game-playing agents. The board median is 27.6%, so Gemini 2.5 Pro Exp (Mar 2025) is 15.7 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
On GPQA Diamond, Gemini 2.5 Pro Exp (Mar 2025) reaches 83.8%. It ranks #87 of 275, ahead of 68% of the field, and 11.9 points behind GPT-6 Astra (max) at 95.8%. The results either side are Kimi K2 Thinking Turbo (84.2%) and openPangu-2.0-Flash (83.7%). It covers expert-level science questions. The board median is 71.7%, so Gemini 2.5 Pro Exp (Mar 2025) is 12.1 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Gemini 2.5 Pro Exp (Mar 2025) manages 66.7% on Fiction.liveBench, good for #15 of 47. grok-4-0709 leads at 94.4%, 27.7 points further up, and 68% of models sit below. The results either side are chatgpt-4o-01-29-2025 (66.7%) and Gemini 2.5 Pro Preview (Mar 2025) (66.7%). It covers long-context story comprehension (16k). The board median is 61.1%, so Gemini 2.5 Pro Exp (Mar 2025) is 5.6 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
SimpleBench puts Gemini 2.5 Pro Exp (Mar 2025) at 51.6%, #20 of 61. That beats 67% of the models tested, but 30.3 points short of Claude Fable 5 (max) (81.9%). The results either side are DeepSeek-V3.2-Speciale (52.6%) and Gemini 2.5 Pro Preview (Mar 2025) (51.6%). It covers trick questions that test common sense. The board median is 41.4%, so Gemini 2.5 Pro Exp (Mar 2025) is 10.2 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
On Humanity's Last Exam, Gemini 2.5 Pro Exp (Mar 2025) reaches 18.2%. It ranks #9 of 27, ahead of 67% of the field, and 22.4 points behind Muse Spark at 40.6%. The results either side are o3 (high) (20.3%) and o4-mini (high) (18.1%). It covers expert-level questions across domains. The board median is 10.7%, so Gemini 2.5 Pro Exp (Mar 2025) is 7.4 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Gemini 2.5 Pro Exp (Mar 2025) manages 4.1% on EnigmaEval, good for #9 of 24. Gemini 3 Pro Preview leads at 18.2%, 14.1 points further up, and 62% of models sit below. The results either side are Gemini 2.5 Pro Preview (Jun 2025) (5.6%) and Kimi K2.5 (3.4%). It covers multi-step puzzle hunts. The board median is 2.9%, so Gemini 2.5 Pro Exp (Mar 2025) is 1.2 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Strengths and weaknesses
Reasoning, maths and science: solid, ahead of 66% of the field on average across 4 benchmarks. Its best board here is GPQA Diamond (#87 of 275) and its weakest is EnigmaEval (#9 of 24).
Release history
The Google DeepMind model released before it was gemini-2.5-pro_32K on 2025-03-25, which scores 49.2. The next one was Gemini 2.5 Pro Preview (Mar 2025) on 2025-03-31, at 50.9.
Reading these numbers
The 6 scores on this page come from 1 source: Epoch AI Benchmarking Hub (6). No official leaderboard supplied any of them, so treat them as third-party measurements. With 6 results the aggregate rank can move a lot when one new score arrives.
Models ranked nearby
- #137 openPangu-2.0-Flash (51.2)
- #138 Claude Opus 4.6 (120K, Low) (51.2)
- #139 Muse-Glimmer-30B (51.2)
- #141 GLM-5.1-FP8 (51.2)
- #142 Llama 3.1-405B (51.2)
- #143 Claude 4.5 Sonnet (51.2)
More from Google DeepMind
- Gemini 3.8 Flash (high) (#26, 54.0)
- Gemini 3.7 Flash (high) (#27, 53.9)
- Gemini 3 Pro Preview (#31, 53.7)
- Gemini 3 Deep Think (#60, 52.6)
- AI co-mathematician (#69, 52.3)
- Gemini 3 Flash (#120, 51.5)
Compare it on the live leaderboard → · How the aggregate score works