Across 2 comparable benchmarks, Gemini 2.5 Deep Think has better reported results on 2 and worse on 0.
Limited evidence2 of 2 shared results countedHow we compare
Price basis: Standard rates. The price section below shows the applicable periods.
| Fact | Gemini 2.5 Deep Think | o3-pro |
|---|---|---|
| Context window | 1M tokens | 200K tokens |
| Input / 1M | $0.400 | $20.00 |
| Output / 1M | $1.60 | $80.00 |
| Blended (4:1) | $0.640 | $32.00 |
| Status | Available | Deprecated |
| Released | August 1, 2025 | June 10, 2025 |
Blended is 4:1 input to output with 90% cache hits at listed rates — set your own mix.
Mean of 1 shared benchmark grouped by capability on a 0–100 scale. How averages are built
Gemini 2.5 Deep Think ahead on 1 of 1 counted
No shared comparable results for Coding, Abstract reasoning, Science & health, Computer & tool use, Research & search, Long context, Multimodal, Professional work & productivity, Cybersecurity.
Published scores, with matching modes where available. Gaps show Gemini 2.5 Deep Think relative to o3-pro. Reading the results
| Benchmark | Gemini 2.5 Deep Think | o3-pro | Gap |
|---|---|---|---|
| AIME2025Mathematics · Accuracy | 99.2Deep Thinking · no tools | 93.0≠Normal · no tools | +6.2 pts#11 vs #29 of 86 |
| HLEKnowledge Exams · Accuracy | 34.8Deep Thinking · no tools | 21.0≠Normal · no tools | +13.8 pts#83 vs #101 of 151 |
A dash means no result is available.
| Benchmark | Gemini 2.5 Deep Think | o3-pro |
|---|---|---|
| AIME 2024 | — | 93.0 |
| FrontierMath | 29.0 | — |
| FrontierMath - Tier 4 | 10.4 | — |
| GPQA Diamond | — | 84.0 |
| IMO-ProofBench | 37.6 | — |
| Aider-Polyglot | — | 84.9 |
| LiveCodeBench | 87.6 | — |
| SWE-bench Verified | — | 75.0 |
| WeirdML v2 | — | 58.2 |
| MCP-Atlas | — | 44.5 |
| ECI | — | 147 |
| Fiction.liveBench | — | 97.2 |
| ARC-AGI-1 | — | 59.3 |
| ARC-AGI-2 | — | 4.9 |
Time-of-day providers use peak-hour rates; fixed-price providers use standard rates. Off-peak discounts remain in the published details.
Drag the handle above the bar for cache hits, the one below it for the input : output mix · 5% steps
Blended price = effective input rate at 90% cache hits × 80% + output rate × 20%
Gemini 2.5 Deep Think costs98%lessthan o3-pro per million tokens — 50× cheaper.Using standard rates at this token mix.
o3-pro: Standard pricing. OpenAI official pricing (developers.openai.com/api/docs/pricing) verified 2026-11-13; standard tier; Deprecated; shutdown 2026-12-11 (replacement: gpt-6.1-sol pro)
| Price / 1M tokens | Gemini 2.5 Deep Think | o3-pro | Ratio |
|---|---|---|---|
| Input | $0.40 | $20.00 | 0.02× |
| Output | $1.60 | $80.00 | 0.02× |
| Cached input | — | — | — |
| Cache write | — | — | — |
| Blended | $0.64 | $32.00 | 0.02× |
Estimates use the selected token mix. Cache writes, tool fees and long-context surcharges are excluded.
Reported cost per task, paired with the score from the same run. Settings and sources may differ; costs do not affect the verdict.
| Benchmark | Gemini 2.5 Deep Think | o3-pro |
|---|---|---|
| ARC-AGI-1 | Not reported | 1 more settings |
| ARC-AGI-2 | Not reported | $7.5516 / task Average · ARC-AGI-2 Score: 4.9 Thinking High (With Tools) $4.7441 / task Average · ARC-AGI-2 Score: 1.9 Thinking Medium (With Tools) |
| Attribute | Gemini 2.5 Deep Think | o3-pro |
|---|---|---|
| Availability | Available | Deprecated |
| Provider | Google Deep Mind | OpenAI |
| Model type | Chat model | Reasoning model |
| First release | August 1, 2025 | June 10, 2025 |
| Context window | 1M tokens | 200K tokens |
| Max output | 16K tokens | 100K tokens |
| Architecture | Undisclosed | Undisclosed |
| Documented capabilities | Reasoning, Multilingual | Reasoning |
| Knowledge cutoff | — | 2024-06-01 |
| Input modalities | Text | Text, Image |
| Output modalities | Text | Text |
| Weights | Closed / API only | Closed / API only |
| Commercial use | Proprietary | Proprietary |
These are reported-score comparisons, not controlled head-to-head tests.
What people found comparing these models, or using either one. Comments also appear on the models' own pages.
Used Gemini 2.5 Deep Think or o3-pro? Sign in with Google to share what worked, what did not, and which API or setup you tested.
No community comments on either model yet.
Alternatives with the most benchmark overlap for this matchup.