Across 14 comparable benchmarks, Gemini 2.5 Pro Experimental 03-25 has better reported results on 11 and worse on 3.
Strong evidence14 of 14 shared results countedHow we compare
Claude Sonnet 4 is no longer available through its official API. API prices for this version are not shown.
Gemini 2.5 Pro Experimental 03-25 is no longer available through its official API. API prices for this version are not shown.
Price basis: Listed rates vs listed rates. The price section below shows the applicable periods.
| Fact | Claude Sonnet 4 | Gemini 2.5 Pro Experimental 03-25 |
|---|---|---|
| Context window | 200K tokens | 1M tokens |
| Input / 1M | — | — |
| Output / 1M | — | — |
| Blended (4:1) | — | — |
| Status | Retired | Retired |
| Released | May 23, 2025 | March 25, 2025 |
Blended is 4:1 input to output with 90% cache hits at listed rates — set your own mix.
Mean of 10 shared benchmarks grouped by capability on a 0–100 scale. How averages are built
Gemini 2.5 Pro Experimental 03-25 ahead on 2 of 3 counted
Gemini 2.5 Pro Experimental 03-25 ahead on 1 of 1 counted
No fixed-scale average — see the rows in Benchmark results
Gemini 2.5 Pro Experimental 03-25 ahead on 3 of 3 counted
Gemini 2.5 Pro Experimental 03-25 ahead on 1 of 1 counted
Split 1–1 of 2 counted
Gemini 2.5 Pro Experimental 03-25 ahead on 1 of 1 counted
Gemini 2.5 Pro Experimental 03-25 ahead on 1 of 1 counted
Claude Sonnet 4 ahead on 1 of 1 counted
No fixed-scale average — see the rows in Benchmark results
No shared comparable results for Research & search, Cybersecurity.
Published scores, with matching modes where available. Gaps show Claude Sonnet 4 relative to Gemini 2.5 Pro Experimental 03-25. Reading the results
| Benchmark | Claude Sonnet 4 | Gemini 2.5 Pro Experimental 03-25 | Gap |
|---|---|---|---|
| Aider-PolyglotAgentic Development · Accuracy | 61.3Thinking · no tools · 32K | 72.9≠Normal · no tools | −11.6 pts#14 vs #8 of 47 |
| LiveCodeBenchAlgorithmic Coding · Pass @K | 48.5Normal · no tools | 70.4Normal · no tools | 0.69×#88 vs #55 of 113 |
| SWE-bench VerifiedRepository Engineering · Accuracy | 80.2Parallel Thinking · tools | 63.8≠Normal · no tools | +16.4 pts#15 vs #75 of 108 |
| SimpleBenchCommonsense · Score (AVG@5) | 45.5Thinking · no tools | 51.6≠Normal · no tools | 0.88×#73 vs #63 of 106 |
| AIME 2024Mathematics · Accuracy | 43.4Normal · no tools | 92.0Normal · no tools | −48.6 pts#45 vs #8 of 54 |
| AIME2025Mathematics · Accuracy | 85.0Deep Thinking · tools | 86.9≠Normal · no tools | −1.9 pts#48 vs #43 of 86 |
| FrontierMath - Tier 4Mathematics · Accuracy | 0.0Normal · no tools | 4.2Normal · no tools | −4.2 pts#44 vs #21 of 50 |
| GPQA DiamondScientific Reasoning · Accuracy | 83.8Deep Thinking · tools | 84.0≠Normal · no tools | −0.2 pts#91 vs #88 of 200 |
| Claw BenchTool Orchestration · Accuracy | 77.8Thinking · tools | 80.4Thinking · tools | −2.6 pts#23 vs #20 of 29 |
| Pinch BenchTool Orchestration · Accuracy | 80.5Thinking · tools | 71.9Thinking · tools | +8.6 pts#23 vs #30 of 38 |
| Fiction.liveBenchLong Reasoning · 16k-token score (%) | 46.9Normal · no tools | 66.7Normal · no tools | −19.8 pts#24 vs #15 of 28 |
| VPCTVisual Understanding · Accuracy (%) | 34.0Thinking · no tools · 32K | 48.0≠Normal · no tools | −14.0 pts#22 vs #8 of 26 |
| Creative WritingWriting | 1483Normal · no tools | 1396Normal · no tools | 1.06×#62 vs #76 of 121 |
| HLEKnowledge Exams · Accuracy | 9.6Thinking · no tools | 18.8≠Normal · no tools | −9.2 pts#130 vs #108 of 151 |
Agentic Development · Accuracy
Algorithmic Coding · Pass @K
Repository Engineering · Accuracy
Commonsense · Score (AVG@5)
Mathematics · Accuracy
Mathematics · Accuracy
Mathematics · Accuracy
Scientific Reasoning · Accuracy
Tool Orchestration · Accuracy
Tool Orchestration · Accuracy
Long Reasoning · 16k-token score (%)
Visual Understanding · Accuracy (%)
Writing
Knowledge Exams · Accuracy
A dash means no result is available.
| Benchmark | Claude Sonnet 4 | Gemini 2.5 Pro Experimental 03-25 |
|---|---|---|
| CritPt | 1.1 | — |
| FrontierMath | 4.1 | — |
| IMO-ProofBench | 27.1 | — |
| IMO-ProofBench Advanced | 4.8 | — |
| MATH-500 | 99.1 | — |
| MMLU-Pro | 84.0 | — |
| CodeClash | 1223 | — |
| IOI 2024 (Vals v1) | 9.7 | — |
| IOI 2025 (Vals v1) | 4 | — |
| SWE-Bench Pro - Commercial | 9.1 | — |
| SWE-Bench Pro - Public | 42.7 | — |
| Terminal Bench Hard | 31.1 | — |
| Terminal-Bench 1.0 | 41.3 | — |
| OSWorld-Verified | 42.2 | — |
| PinchBench v2 | 48.8 | — |
| ECI | 142 | — |
| LiveBench | 61.3 | — |
| IF Bench | 55.0 | — |
| GeoBench ACW | — | 81.0 |
| MMMU | 74.4 | — |
| MMMU-Pro | 62.4 | — |
| GDPval-AA | 33.0 | — |
| SAGE | 35.0 | — |
| τ²-Bench | 52.0 | — |
| τ³-Banking | 16.7 | — |
| ARC-AGI-1 | 40.0 | — |
| ARC-AGI-2 | 5.9 | — |
| BALROG | — | 43.3 |
| SimpleQA | — | 52.9 |
| MedCode | 35.0 | — |
| MedScribe | 72.4 | — |
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%
A blended cost needs a published input and output price on both sides, and one of these two does not have them yet.
Claude Sonnet 4: Listed text rates; pricing tier not confirmed.
Gemini 2.5 Pro Experimental 03-25: Listed text rates; pricing tier not confirmed.
| Price / 1M tokens | Claude Sonnet 4 | Gemini 2.5 Pro Experimental 03-25 | Ratio |
|---|---|---|---|
| Input | — | — | — |
| Output | — | — | — |
| Cached input | — | — | — |
| Cache write | — | — | — |
| Blended | — | — | — |
Estimates use the selected token mix. Cache writes, tool fees and long-context surcharges are excluded.
Lower prices are highlighted. The cache hit rate applies to input tokens only; models without a published cached-input rate use their normal input price for those tokens.Effective dates: Claude Sonnet 4 2026-11-14; Gemini 2.5 Pro Experimental 03-25 —.Reported cost per task, paired with the score from the same run. Settings and sources may differ; costs do not affect the verdict.
| Attribute | Claude Sonnet 4 | Gemini 2.5 Pro Experimental 03-25 |
|---|---|---|
| Availability | Retired | Retired |
| Provider | Anthropic | Google Deep Mind |
| Model type | Reasoning model | Reasoning model |
| First release | May 23, 2025 | March 25, 2025 |
| Context window | 200K tokens | 1M tokens |
| Max output | 64K tokens | 64K tokens |
| Architecture | Undisclosed | Undisclosed |
| Documented capabilities | Reasoning, Multilingual | Reasoning, Multilingual |
| Input modalities | Text, Image | Text, Image, Audio, Video |
| 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 Claude Sonnet 4 or Gemini 2.5 Pro Experimental 03-25? 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.