Model comparison
Claude Haiku 4.5 vs Gemini 3 Pro
Gemini 3 Pro is the stronger model overall, scoring 54.8 to 39.5 on the Noometry Index.
Last verified . 45 shared benchmarks.
GoogleSummary
- They share 45 benchmarks with published results for both. Claude Haiku 4.5 scores higher in 0 categories and Gemini 3 Pro in 10 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3 Pro leads 52.5 to 15.1.
- The biggest single-benchmark swing is NYT Connections (extended): 14.3% for Claude Haiku 4.5 and 94.4% for Gemini 3 Pro.
Side by side
| Claude Haiku 4.5 | Gemini 3 Pro | |
|---|---|---|
| Provider | Anthropic | |
| Noometry Index | 39.5 | 54.8 |
| Released | 2025-10-15 | 2025-11-18 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | — |
| Max output | 64K | — |
| Input $ / M tokens | $1 | — |
| Output $ / M tokens | $5 | — |
| Results tracked | 53 | 67 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 3 Pro leads
Claude Haiku 4.5: 44.0 (#78), Gemini 3 Pro: 51.6 (#39)
| Benchmark | Claude Haiku 4.5 | Gemini 3 Pro |
|---|---|---|
| SWE-bench Verified (bash only) | 66.6% | 74.2% |
| LMArena WebDev | 1330 | 1440 |
| SWE-bench Multilingual | 64.7% | 68.7% |
| WeirdML | 45.4% | 69.9% |
| LMArena Coding | 1453 | 1481 |
| ALE-Bench | 653.48 | 1,177 |
| SWE-bench Verified | — | 72.9% |
| SciCode | 43.3% | — |
| GSO | — | 18.6% |
| AlgoTune | — | 1.83 |
Agentic & Tool Use Gemini 3 Pro leads
Claude Haiku 4.5: 33.6 (#52), Gemini 3 Pro: 40.6 (#23)
| Benchmark | Claude Haiku 4.5 | Gemini 3 Pro |
|---|---|---|
| Terminal-Bench | 35.5% | 69.4% |
| Berkeley Function Calling Leaderboard | 68.7% | 72.5% |
| DeepResearch Bench | 45.5% | 46.3% |
| BALROG | 31.2% | 58.1% |
| Vending-Bench 2 | 458.89 | 5,478 |
| GDPval | — | 40.3% |
| Remote Labor Index | — | 1.3% |
| τ²-bench Airline | — | 80.5% |
| τ²-bench Banking | — | 18% |
| τ²-bench Retail | — | 75.9% |
| τ²-bench Telecom | — | 91% |
| ExploitBench | 13.7% | — |
| LMArena Search | — | 1207 |
| METR Time Horizons | — | 71% |
Reasoning Gemini 3 Pro leads
Claude Haiku 4.5: 15.1 (#320), Gemini 3 Pro: 52.5 (#31)
| Benchmark | Claude Haiku 4.5 | Gemini 3 Pro |
|---|---|---|
| ARC-AGI-2 | 4% | 31.1% |
| NYT Connections (extended) | 14.3% | 94.4% |
| ARC-AGI-1 | 47.7% | 75% |
| CritPt | 0% | 6.9% |
| Chess Puzzles | 8% | 31% |
| LMArena Hard Prompts | 1420 | 1480 |
| Epoch Capabilities Index | 142.41 | 152.92 |
| ForecastBench | 61.4 | 61.2 |
| SimpleBench | — | 76.4% |
| Kagi LLM Benchmark | — | 80.1% |
| EnigmaEval | — | 18.2% |
| DTBench | 73.6% | — |
| LMCA | 30.9% | — |
Math Gemini 3 Pro leads
Claude Haiku 4.5: 44.9 (#78), Gemini 3 Pro: 49.9 (#59)
| Benchmark | Claude Haiku 4.5 | Gemini 3 Pro |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 66.7% | 91.4% |
| Omni-MATH | 56.1% | 55.5% |
| LMArena Math | 1396 | 1476 |
| FrontierMath (Feb 2025 set) | 5.9% | 37.6% |
| FrontierMath Tier 4 (v1) | 2.1% | 18.8% |
| MathArena Final-Answer Competitions | — | 67% |
| ProofBench | — | 20% |
| MATH Level 5 | 96.4% | — |
Knowledge Gemini 3 Pro leads
Claude Haiku 4.5: 37.7 (#153), Gemini 3 Pro: 64.4 (#16)
| Benchmark | Claude Haiku 4.5 | Gemini 3 Pro |
|---|---|---|
| GPQA Diamond | 71.2% | 92.6% |
| MMLU-Pro | 77.7% | 90.3% |
| Vectara Hallucination Rate | 9.8% | 13.6% |
| GPQA (HELM) | 60.5% | 80.3% |
| LMArena Expert | 1442 | 1475 |
| Humanity's Last Exam | — | 37.5% |
| SimpleQA Verified | 13.2% | — |
Multimodal Gemini 3 Pro leads
Claude Haiku 4.5: 26.8 (#118), Gemini 3 Pro: 57.6 (#2)
| Benchmark | Claude Haiku 4.5 | Gemini 3 Pro |
|---|---|---|
| LMArena Document | 1420 | 1434 |
| LMArena Vision | — | 1305 |
| GeoBench | — | 84% |
| VPCT | — | 91% |
| Blueprint-Bench 2 | 0% | — |
Multilingual Gemini 3 Pro leads
Claude Haiku 4.5: 49.9 (#129), Gemini 3 Pro: 56.9 (#16)
| Benchmark | Claude Haiku 4.5 | Gemini 3 Pro |
|---|---|---|
| LMArena Non-English | 1377 | 1474 |
| LMArena Chinese | 1417 | 1523 |
| LMArena French | 1408 | 1492 |
| LMArena German | 1375 | 1515 |
| LMArena Japanese | 1339 | 1510 |
| LMArena Korean | 1347 | 1448 |
| LMArena Russian | 1381 | 1493 |
| LMArena Spanish | 1420 | 1470 |
Instruction Following Gemini 3 Pro leads
Claude Haiku 4.5: 71.4 (#149), Gemini 3 Pro: 76.3 (#45)
| Benchmark | Claude Haiku 4.5 | Gemini 3 Pro |
|---|---|---|
| IFEval | 80.1% | 87.7% |
| LMArena Instruction Following | 1414 | 1458 |
Long Context Too close to call
Claude Haiku 4.5: 43.6 (#92), Gemini 3 Pro: 44.0 (#79)
| Benchmark | Claude Haiku 4.5 | Gemini 3 Pro |
|---|---|---|
| LMArena Longer Query | 1427 | 1471 |
| CL-bench | — | 15.8% |
Writing & Preference Gemini 3 Pro leads
Claude Haiku 4.5: 57.9 (#123), Gemini 3 Pro: 66.4 (#35)
| Benchmark | Claude Haiku 4.5 | Gemini 3 Pro |
|---|---|---|
| LMArena Text | 1396 | 1479 |
| LMArena Creative Writing | 1372 | 1482 |
| WildBench | 83.9% | 85.9% |
| LMArena Multi-Turn | 1409 | 1484 |
| EQ-Bench Creative Writing | — | 1525 |
| EQ-Bench 4 | 1064 | — |
Frequently asked questions
Is Claude Haiku 4.5 better than Gemini 3 Pro?
Gemini 3 Pro is the stronger model overall, scoring 54.8 to 39.5 on the Noometry Index.
Is Claude Haiku 4.5 or Gemini 3 Pro better for coding?
Gemini 3 Pro scores higher on coding benchmarks: 51.6 versus 44.0 in the Noometry coding category.
How many benchmarks do Claude Haiku 4.5 and Gemini 3 Pro share?
45 benchmarks have published results for both models. Claude Haiku 4.5 has 53 scored results on Noometry and Gemini 3 Pro has 67.