Across 11 comparable benchmarks, OpenAI o3 has better reported results on 11 and worse on 0.
Moderate evidence11 of 11 shared results countedHow we compare
Price basis: Standard rates. The price section below shows the applicable periods.
| Fact | Gemini 2.5 Flash-Lite | OpenAI o3 |
|---|---|---|
| Context window | 1M tokens | 200K tokens |
| Input / 1M | $0.100 | $2.00 |
| Output / 1M | $0.400 | $8.00 |
| Blended (4:1) | $0.095 | $2.12 |
| Status | Available | Deprecated |
| Released | June 17, 2025 | April 16, 2025 |
Blended is 4:1 input to output with 90% cache hits at listed rates — set your own mix.
Mean of 8 shared benchmarks grouped by capability on a 0–100 scale. How averages are built
OpenAI o3 ahead on 3 of 3 counted
OpenAI o3 ahead on 1 of 1 counted
OpenAI o3 ahead on 3 of 3 counted
OpenAI o3 ahead on 1 of 1 counted
OpenAI o3 ahead on 1 of 1 counted
No shared comparable results for Abstract reasoning, Computer & tool use, Research & search, Long context, Cybersecurity.
Published scores, with matching modes where available. Gaps show Gemini 2.5 Flash-Lite relative to OpenAI o3. Reading the results
| Benchmark | Gemini 2.5 Flash-Lite | OpenAI o3 | Gap |
|---|---|---|---|
| LiveCodeBenchAlgorithmic Coding · Pass @K | 34.3Normal · no tools | 75.8Normal · no tools | 0.45×#102 vs #43 of 113 |
| SWE-bench VerifiedRepository Engineering · Accuracy | 27.6Normal · no tools | 69.1≠Thinking · no tools | −41.5 pts#105 vs #65 of 108 |
| Terminal Bench HardAgentic Development · Accuracy | 4.5Thinking · tools | 37.1Thinking · tools | −32.6 pts#137 vs #43 of 157 |
| AIME2025Mathematics · Accuracy | 63.1Normal · no tools | 88.9≠Thinking · no tools | −25.8 pts#73 vs #37 of 86 |
| GPQA DiamondScientific Reasoning · Accuracy | 66.7Normal · no tools | 83.3≠Thinking · no tools | −16.6 pts#151 vs #94 of 200 |
| MedCodeClinical Workflows · Accuracy (%) | 27.1Unspecified Reasoning · tools | 47.3≠Thinking High · tools | −20.2 pts#93 vs #33 of 97 |
| MedScribeClinical Workflows · Rubric pass rate (%) | 72.8Unspecified Reasoning · tools | 76.7≠Thinking High · tools | −3.8 pts#79 vs #67 of 99 |
| MMMU-ProVisual Understanding · Accuracy | 58.2Thinking · no tools | 76.4≠Reported best | −18.2 pts#110 vs #56 of 136 |
| SAGEService Workflows · Accuracy (%) | 27.5Unspecified Reasoning · tools | 41.8≠Thinking High · tools | −14.2 pts#79 vs #56 of 82 |
| Reported separately from the capability domains; not included in the averages. Read each metric’s direction — lower can be better. | |||
| SimpleQAHonesty & Factuality · Accuracy | 13.0Normal · no tools | 49.4Normal · no tools | −36.4 pts#38 vs #14 of 46 |
| HLEKnowledge Exams · Accuracy | 6.9Normal · no tools | 20.3≠Thinking High · no tools | −13.4 pts#137 vs #102 of 151 |
Algorithmic Coding · Pass @K
Repository Engineering · Accuracy
Agentic Development · Accuracy
Mathematics · Accuracy
Scientific Reasoning · Accuracy
Clinical Workflows · Accuracy (%)
Clinical Workflows · Rubric pass rate (%)
Visual Understanding · Accuracy
Service Workflows · Accuracy (%)
Honesty & Factuality · Accuracy
Knowledge Exams · Accuracy
A dash means no result is available.
| Benchmark | Gemini 2.5 Flash-Lite | OpenAI o3 |
|---|---|---|
| AIME 2024 | — | 91.6 |
| CritPt | — | 1.1 |
| FrontierMath | — | 10.3 |
| FrontierMath - Tier 4 | — | 2.1 |
| FrontierMath v2 | — | 33.3 |
| IMO-ProofBench | — | 20.5 |
| IMO-ProofBench Advanced | — | 20.5 |
| MATH-500 | — | 99.2 |
| MMLU | 84.5 | — |
| MMLU-Pro | — | 85.6 |
| Aider-Polyglot | — | 81.3 |
| CodeClash | — | 1343 |
| GSO | — | 8.8 |
| Terminal-Bench 1.0 | — | 30.2 |
| WeirdML v2 | — | 52.4 |
| OSWorld-Verified | — | 23.0 |
| ECI | — | 147 |
| LiveBench | 42.6 | — |
| METR Time Horizons v1.1 | — | 91.3 |
| Creative Writing | — | 1675 |
| Fiction.liveBench | — | 88.9 |
| GeoBench ACW | — | 74.0 |
| MMMU | — | 82.9 |
| VPCT | — | 52.0 |
| ARC-AGI-1 | — | 60.8 |
| ARC-AGI-2 | — | 6.5 |
| SimpleBench | — | 53.1 |
| GAIA | — | 28.5 |
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 Flash-Lite costs96%lessthan OpenAI o3 per million tokens — 22.3× cheaper.Using standard rates at this token mix.
Gemini 2.5 Flash-Lite: Standard pricing. Google Gemini API official pricing (ai.google.dev/gemini-api/docs/pricing) verified 2026-11-13; paid tier; standard tier; text input Google Gemini API official pricing (ai.google.dev/gemini-api/docs/pricing) verified 2026-11-13; paid tier; standard tier; output
OpenAI o3: 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)
| Price / 1M tokens | Gemini 2.5 Flash-Lite | OpenAI o3 | Ratio |
|---|---|---|---|
| Input | $0.10 | $2.00 | 0.05× |
| Output | $0.40 | $8.00 | 0.05× |
| Cached input | $0.01 | $0.50 | 0.02× |
| Cache write | — | — | — |
| Blended | $0.095 | $2.12 | 0.04× |
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.
| Attribute | Gemini 2.5 Flash-Lite | OpenAI o3 |
|---|---|---|
| Availability | Available | Deprecated |
| Provider | Google Deep Mind | OpenAI |
| Model type | Reasoning model | Reasoning model |
| First release | June 17, 2025 | April 16, 2025 |
| Context window | 1M tokens | 200K tokens |
| Max output | 64K tokens | 100K tokens |
| Architecture | Undisclosed | Undisclosed |
| Documented capabilities | Reasoning, Multilingual | Reasoning, Multilingual |
| Input modalities | Text, Image, Audio, Video | 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 Flash-Lite or OpenAI o3? 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.