Gemini 2.5 Pro Preview (Jun 2025): benchmark scores, price and ranking
Gemini 2.5 Pro Preview (Jun 2025) from Google DeepMind ranks #147 of 743 LLMs on the LLMs Tiger aggregate benchmark score (51.0), based on 10 benchmark results captured up to 2026-10-13.
- Organisation
- Google DeepMind
- Type
- Preview release
- Open weights
- No
- Released
- 2025-05-06 (source: epoch)
- Cheapest API price
- $1.25 in / $10.00 out per million tokens (via deepinfra, observed 2026-10-13T19:47Z)
- Aggregate rank
- #147 of 743 (score 51.0)
Relative to other models, Gemini 2.5 Pro Preview (Jun 2025) is strongest on Fiction.liveBench (#3 of 47) and weakest on GSO-Bench (#13 of 19).
Across the 8 benchmarks where at least ten models are scored, Gemini 2.5 Pro Preview (Jun 2025) is in the top 10% on 2 and in the top half on 7.
Its cheapest listed input price, $1.25 per million tokens, is lower than that of 26% of the 427 models with a listed price.
Gemini 2.5 Pro Preview (Jun 2025) benchmark results
| Benchmark | Score | Rank | Source | Captured |
|---|---|---|---|---|
| Fiction.liveBench | 91.7% | #3 of 47 | Epoch AI Benchmarking Hub | 2026-10-13 |
| SimpleBench | 62.4% | #5 of 61 | Epoch AI Benchmarking Hub | 2026-10-13 |
| Humanity's Last Exam | 21.6% | #7 of 27 | Epoch AI Benchmarking Hub | 2026-10-13 |
| GPQA Diamond | 84.8% | #82 of 275 | Epoch AI Benchmarking Hub | 2026-10-13 |
| TheAgentCompany | 30.3% | #4 of 13 | Epoch AI Benchmarking Hub | 2026-10-13 |
| EnigmaEval | 5.6% | #8 of 24 | Epoch AI Benchmarking Hub | 2026-10-13 |
| METR time horizon | 39 min | #13 of 31 | Epoch AI Benchmarking Hub | 2026-10-13 |
| GSO-Bench | 3.9% | #13 of 19 | Epoch AI Benchmarking Hub | 2026-10-13 |
| DeepResearch Bench | 42.8% | #6 of 7 | Epoch AI Benchmarking Hub | 2026-10-13 |
| Remote Labor Index | 0.8% | #2 of 2 | Epoch AI Benchmarking Hub | 2026-10-13 |
About Gemini 2.5 Pro Preview (Jun 2025)
Gemini 2.5 Pro Preview (Jun 2025) is a preview release from Google DeepMind. It was released on 2025-05-06, according to Epoch AI. The cheapest API listing we track charges $1.25 per million input tokens and $10.00 per million output tokens, through deepinfra. We hold 10 benchmark results for it across 4 areas, and 2 are in the top tenth of their boards.
Where Gemini 2.5 Pro Preview (Jun 2025) stands
On our aggregate score, Gemini 2.5 Pro Preview (Jun 2025) is #147 of 743 ranked models, in the top quarter. Its score is 51.0; the scale is centred on 50, so it sits 1.0 points above the typical model. That beats 80% of the ranked field. Among the 41 ranked Google DeepMind models it is #8. The next one up is Gemini 2.5 Pro Exp (Mar 2025) at 51.2. The next one down is Gemini 2.5 Pro Preview (Mar 2025) at 50.9.
Gemini 2.5 Pro Preview (Jun 2025), benchmark by benchmark
On Fiction.liveBench, Gemini 2.5 Pro Preview (Jun 2025) scores 91.7% and ranks #3 of 47. Only 2 models score higher, led by grok-4-0709 at 94.4%, which is 2.7 points ahead. It covers long-context story comprehension (16k). The board median is 61.1%, so Gemini 2.5 Pro Preview (Jun 2025) is 30.6 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
On SimpleBench, Gemini 2.5 Pro Preview (Jun 2025) scores 62.4% and ranks #5 of 61. Only 4 models score higher, led by Claude Fable 5 (max) at 81.9%, which is 19.5 points ahead. It covers trick questions that test common sense. The board median is 41.4%, so Gemini 2.5 Pro Preview (Jun 2025) is 21.0 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Gemini 2.5 Pro Preview (Jun 2025) manages 21.6% on Humanity's Last Exam, good for #7 of 27. Muse Spark leads at 40.6%, 18.9 points further up, and 74% of models sit below. The results either side are Kimi K2.5 (24.4%) and o3 (high) (20.3%). It covers expert-level questions across domains. The board median is 10.7%, so Gemini 2.5 Pro Preview (Jun 2025) is 10.9 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
GPQA Diamond puts Gemini 2.5 Pro Preview (Jun 2025) at 84.8%, #82 of 275. That beats 70% of the models tested, but 10.9 points short of GPT-6 Astra (max) (95.8%). The results either side are Darwin-4B-David (85.0%) and Qwen3.5 Plus (84.8%). It covers expert-level science questions. The board median is 71.7%, so Gemini 2.5 Pro Preview (Jun 2025) is 13.1 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
On TheAgentCompany, Gemini 2.5 Pro Preview (Jun 2025) reaches 30.3%. It ranks #4 of 13, ahead of 69% of the field, and 10.8 points behind gemini-2.5-flash-preview-09-2025 at 41.1%. The results either side are claude-3-7-sonnet-20250219 (30.9%) and Claude 3.5 Sonnet (Oct 2024) (24.0%). It covers work tasks in a simulated software company. The board median is 8.6%, so Gemini 2.5 Pro Preview (Jun 2025) is 21.7 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Gemini 2.5 Pro Preview (Jun 2025) manages 5.6% on EnigmaEval, good for #8 of 24. Gemini 3 Pro Preview leads at 18.2%, 12.7 points further up, and 67% of models sit below. The results either side are Claude Opus 4 (5.6%) and Gemini 2.5 Pro Exp (Mar 2025) (4.1%). It covers multi-step puzzle hunts. The board median is 2.9%, so Gemini 2.5 Pro Preview (Jun 2025) is 2.7 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
METR time horizon puts Gemini 2.5 Pro Preview (Jun 2025) at 39 min, #13 of 31. That beats 58% of the models tested, but 1006 minutes short of Claude Mythos Preview (Early) (1,045 min). The results either side are Kimi K2 Thinking (54 min) and DeepSeek-R1 (May 2025) (31 min). It covers length of task (minutes) done at 50% success. The board median is 23 min, so Gemini 2.5 Pro Preview (Jun 2025) is 16 minutes above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Gemini 2.5 Pro Preview (Jun 2025) lands below the middle on GSO-Bench: 3.9%, #13 of 19, with 63% of the field scoring higher. GPT-5.5 (xhigh) leads at 40.2%; the closest result above is Claude 3.5 Sonnet (Oct 2024) at 4.6%. It covers software performance optimisation. The board median is 4.9%, so Gemini 2.5 Pro Preview (Jun 2025) is 1.0 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
The full table above also lists the other 2 results, including boards with fewer than ten models, where a rank says little.
Strengths and weaknesses
Reasoning, maths and science: strong, ahead of 76% of the field on average across 4 benchmarks. Its best board here is SimpleBench (#5 of 61) and its weakest is EnigmaEval (#8 of 24). General knowledge and chat: strong, ahead of 76% of the field on average across 2 benchmarks. Its best board here is Fiction.liveBench (#3 of 47) and its weakest is METR time horizon (#13 of 31).
Price and value
The median input price across the 427 priced models is $0.500 per million tokens, so Gemini 2.5 Pro Preview (Jun 2025) costs more than the typical listing. Models that score within 1.5 points or better and cost less: Qwen3.7 Flash ($0.030, 50.5), GLM-5.2 (high) ($0.060, 51.6), glm-4.7-flash ($0.060, 50.1). Similar scores cost more on glm-5p1 ($1.40) and gemini-3.5-flash-high ($1.50).
Release history
The Google DeepMind model released before it was Gemini 2.5 Flash Preview (Apr 2025) on 2025-04-17, which scores 48.6. The next one was gemini-2.5-flash-preview-05-20 on 2025-05-20, at 50.5.
Reading these numbers
The 10 scores on this page come from 1 source: Epoch AI Benchmarking Hub (10). No official leaderboard supplied any of them, so treat them as third-party measurements.
Models ranked nearby
- #144 GPT-5 (high) (51.1)
- #145 Claude Code + Fable 5 (51.1)
- #146 Qwen3.5 397B-A17B (51.1)
- #148 grok-4-1-fast-non-reasoning (51.0)
- #149 Quasar Alpha (51.0)
- #150 ERNIE-4.5-300B-A47B-PT (51.0)
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