Gemini 2.5 Flash Preview (Apr 2025): benchmark scores, price and ranking
Gemini 2.5 Flash Preview (Apr 2025) from Google DeepMind ranks #340 of 743 LLMs on the LLMs Tiger aggregate benchmark score (48.6), based on 3 benchmark results captured up to 2026-10-13.
- Organisation
- Google DeepMind
- Type
- Preview release
- Open weights
- No
- Released
- 2025-04-17 (source: epoch)
- Cheapest API price
- $0.300 in / $2.50 out per million tokens (via deepinfra, observed 2026-10-13T20:47Z)
- Aggregate rank
- #340 of 743 (score 48.6)
Relative to other models, Gemini 2.5 Flash Preview (Apr 2025) is strongest on OTIS Mock AIME (#74 of 189) and weakest on Fiction.liveBench (#35 of 47).
Across the 3 benchmarks where at least ten models are scored, Gemini 2.5 Flash Preview (Apr 2025) is in the top 10% on 0 and in the top half on 2.
Its cheapest listed input price, $0.300 per million tokens, is lower than that of 58% of the 427 models with a listed price.
Gemini 2.5 Flash Preview (Apr 2025) benchmark results
| Benchmark | Score | Rank | Source | Captured |
|---|---|---|---|---|
| OTIS Mock AIME | 73.1% | #74 of 189 | Epoch AI Benchmarking Hub | 2026-10-13 |
| Humanity's Last Exam | 12.1% | #11 of 27 | Epoch AI Benchmarking Hub | 2026-10-13 |
| Fiction.liveBench | 47.2% | #35 of 47 | Epoch AI Benchmarking Hub | 2026-10-13 |
About Gemini 2.5 Flash Preview (Apr 2025)
Gemini 2.5 Flash Preview (Apr 2025) is a preview release from Google DeepMind. It was released on 2025-04-17, according to Epoch AI. The cheapest API listing we track charges $0.300 per million input tokens and $2.50 per million output tokens, through deepinfra. We hold 3 benchmark results for it across 2 areas, and none is in the top tenth of its board.
Where Gemini 2.5 Flash Preview (Apr 2025) stands
On our aggregate score, Gemini 2.5 Flash Preview (Apr 2025) is #340 of 743 ranked models, around the middle of the table. Its score is 48.6; the scale is centred on 50, so it sits 1.4 points below the typical model. That beats 54% of the ranked field. Among the 41 ranked Google DeepMind models it is #19. The next one up is gemini-2.5-flash-lite-preview-06-17_16K at 48.7. The next one down is gemini-1.5-pro-001-feb24 at 48.1.
Gemini 2.5 Flash Preview (Apr 2025), benchmark by benchmark
OTIS Mock AIME puts Gemini 2.5 Flash Preview (Apr 2025) at 73.1%, #74 of 189. That beats 61% of the models tested, but 26.9 points short of Claude Fable 5.1 (max) (100.0%). The results either side are Qwen3-Max-Instruct (73.3%) and Claude Sonnet 4.6 (max) (71.1%). It covers competition maths (AIME 2024-25 style). The board median is 57.8%, so Gemini 2.5 Flash Preview (Apr 2025) is 15.3 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
On Humanity's Last Exam, Gemini 2.5 Flash Preview (Apr 2025) reaches 12.1%. It ranks #11 of 27, ahead of 59% of the field, and 28.5 points behind Muse Spark at 40.6%. The results either side are o4-mini (high) (18.1%) and Claude Opus 4.1 (unknown thinking) (11.5%). It covers expert-level questions across domains. The board median is 10.7%, so Gemini 2.5 Flash Preview (Apr 2025) is 1.4 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Gemini 2.5 Flash Preview (Apr 2025) lands below the middle on Fiction.liveBench: 47.2%, #35 of 47, with 72% of the field scoring higher. grok-4-0709 leads at 94.4%; the closest result above is DeepSeek-V3 (Mar 2025) at 50.0%. It covers long-context story comprehension (16k). The board median is 61.1%, so Gemini 2.5 Flash Preview (Apr 2025) is 13.9 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Strengths and weaknesses
Reasoning, maths and science: solid, ahead of 60% of the field on average across 2 benchmarks. Its best board here is OTIS Mock AIME (#74 of 189) and its weakest is Humanity's Last Exam (#11 of 27).
Price and value
The median input price across the 427 priced models is $0.500 per million tokens, so Gemini 2.5 Flash Preview (Apr 2025) costs less 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), DeepSeek V4 Flash 0731 (max) ($0.016, 53.1). Similar scores cost more on Devstral (2512) ($0.400) and GLM-4.5 Thinking ($0.400).
Release history
The Google DeepMind model released before it was gemini-2.5-flash on 2025-04-17, which scores 45.6. The next one was Gemini 2.5 Pro Preview (Jun 2025) on 2025-05-06, at 51.0.
Reading these numbers
The 3 scores on this page come from 1 source: Epoch AI Benchmarking Hub (3). No official leaderboard supplied any of them, so treat them as third-party measurements. With 3 results the aggregate rank can move a lot when one new score arrives.
Models ranked nearby
- #337 Llama3-SWE-RL-70B (48.6)
- #338 grok-4.20-multi-agent-beta-0309 (48.6)
- #339 Qwen3.7 Plus (48.6)
- #341 Seed-OSS-36B-Base-woSyn (48.6)
- #342 qwen2.5-coder:7b-instruct-q8_0 (48.5)
- #343 QwQ-32B + Qwen 2.5 Coder Instruct (48.5)
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