Model comparison
Claude Haiku 4.5 vs o1
o1 is the stronger model overall, scoring 40.9 to 39.5 on the Noometry Index. Claude Haiku 4.5 costs 13× less per token, which makes it the better buy when o1's lead doesn't matter for your workload.
Last verified . 28 shared benchmarks.
Summary
- They share 28 benchmarks with published results for both. Claude Haiku 4.5 scores higher in 4 categories and o1 in 6 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where o1 leads 27.9 to 15.1.
- The biggest single-benchmark swing is SimpleQA Verified: 13.2% for Claude Haiku 4.5 and 41.1% for o1.
- Claude Haiku 4.5 is cheaper at $1 / $5 per million input/output tokens, against $15 / $60 for o1.
Side by side
| Claude Haiku 4.5 | o1 | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 39.5 | 40.9 |
| Released | 2025-10-15 | 2024-09-12 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | 200K |
| Max output | 64K | 100K |
| Input $ / M tokens | $1 | $15 |
| Output $ / M tokens | $5 | $60 |
| Results tracked | 53 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o1 leads
Claude Haiku 4.5: 44.0 (#78), o1: 46.1 (#70)
| Benchmark | Claude Haiku 4.5 | o1 |
|---|---|---|
| WeirdML | 45.4% | 47.6% |
| LMArena Coding | 1453 | 1367 |
| SWE-bench Verified (bash only) | 66.6% | — |
| Aider Polyglot | — | 61.7% |
| LMArena WebDev | 1330 | — |
| SWE-bench Multilingual | 64.7% | — |
| SciCode | 43.3% | — |
| LiveBench Coding | — | 69.7% |
| CadEval | — | 56% |
| ALE-Bench | 653.48 | — |
| HumanEval+ | — | 89% |
| MBPP+ | — | 80.2% |
Agentic & Tool Use Claude Haiku 4.5 leads
Claude Haiku 4.5: 33.6 (#52), o1: 24.6 (#117)
| Benchmark | Claude Haiku 4.5 | o1 |
|---|---|---|
| Terminal-Bench | 35.5% | — |
| Berkeley Function Calling Leaderboard | 68.7% | — |
| Cybench | — | 10% |
| DeepResearch Bench | 45.5% | — |
| BALROG | 31.2% | — |
| ExploitBench | 13.7% | — |
| METR Time Horizons | — | 51.1% |
| Vending-Bench 2 | 458.89 | — |
Reasoning o1 leads
Claude Haiku 4.5: 15.1 (#320), o1: 27.9 (#111)
| Benchmark | Claude Haiku 4.5 | o1 |
|---|---|---|
| ARC-AGI-1 | 47.7% | 30.7% |
| Chess Puzzles | 8% | 15% |
| LMArena Hard Prompts | 1420 | 1371 |
| DTBench | 73.6% | 74.7% |
| LMCA | 30.9% | 22.3% |
| Epoch Capabilities Index | 142.41 | 141.91 |
| ARC-AGI-2 | 4% | — |
| SimpleBench | — | 41.7% |
| NYT Connections (extended) | 14.3% | — |
| CritPt | 0% | — |
| EnigmaEval | — | 5.7% |
| LiveBench Reasoning | — | 91.6% |
| LiveBench Data Analysis | — | 65.5% |
| ForecastBench | 61.4 | — |
| LiveBench | — | 75.7% |
Math Claude Haiku 4.5 leads
Claude Haiku 4.5: 44.9 (#78), o1: 36.1 (#175)
| Benchmark | Claude Haiku 4.5 | o1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 66.7% | 73.3% |
| LMArena Math | 1396 | 1388 |
| MATH Level 5 | 96.4% | 94.7% |
| FrontierMath (Feb 2025 set) | 5.9% | 9.3% |
| FrontierMath (Tiers 1-3) | — | 14.7% |
| Omni-MATH | 56.1% | — |
| LiveBench Math | — | 80.3% |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge o1 leads
Claude Haiku 4.5: 37.7 (#153), o1: 41.5 (#110)
| Benchmark | Claude Haiku 4.5 | o1 |
|---|---|---|
| GPQA Diamond | 71.2% | 76.8% |
| SimpleQA Verified | 13.2% | 41.1% |
| LMArena Expert | 1442 | 1361 |
| Humanity's Last Exam | — | 8% |
| MMLU-Pro | 77.7% | — |
| Confabulations | — | 11.7% |
| Vectara Hallucination Rate | 9.8% | — |
| GPQA (HELM) | 60.5% | — |
Multimodal o1 leads
Claude Haiku 4.5: 26.8 (#118), o1: 34.2 (#93)
| Benchmark | Claude Haiku 4.5 | o1 |
|---|---|---|
| LMArena Vision | — | 1168 |
| GeoBench | — | 80% |
| VPCT | — | 37% |
| Blueprint-Bench 2 | 0% | — |
| LMArena Document | 1420 | — |
| SpatialViz-Bench | — | 41.4% |
Multilingual Claude Haiku 4.5 leads
Claude Haiku 4.5: 49.9 (#129), o1: 48.6 (#142)
| Benchmark | Claude Haiku 4.5 | o1 |
|---|---|---|
| LMArena Non-English | 1377 | 1358 |
| LMArena Chinese | 1417 | 1394 |
| LMArena French | 1408 | 1344 |
| LMArena German | 1375 | 1337 |
| LMArena Japanese | 1339 | 1346 |
| LMArena Korean | 1347 | 1396 |
| LMArena Russian | 1381 | 1356 |
| LMArena Spanish | 1420 | 1345 |
Instruction Following o1 leads
Claude Haiku 4.5: 71.4 (#149), o1: 74.8 (#86)
| Benchmark | Claude Haiku 4.5 | o1 |
|---|---|---|
| LMArena Instruction Following | 1414 | 1367 |
| LiveBench Instruction Following | — | 81.5% |
| IFEval | 80.1% | — |
Long Context o1 leads
Claude Haiku 4.5: 43.6 (#92), o1: 50.3 (#9)
| Benchmark | Claude Haiku 4.5 | o1 |
|---|---|---|
| LMArena Longer Query | 1427 | 1378 |
| Fiction.LiveBench | — | 83.3% |
Writing & Preference Claude Haiku 4.5 leads
Claude Haiku 4.5: 57.9 (#123), o1: 55.6 (#144)
| Benchmark | Claude Haiku 4.5 | o1 |
|---|---|---|
| LMArena Text | 1396 | 1366 |
| LMArena Creative Writing | 1372 | 1348 |
| LMArena Multi-Turn | 1409 | 1369 |
| Short-Story Creative Writing | — | 70.2% |
| WildBench | 83.9% | — |
| EQ-Bench 4 | 1064 | — |
| LiveBench Language | — | 65.4% |
Frequently asked questions
Is Claude Haiku 4.5 better than o1?
o1 is the stronger model overall, scoring 40.9 to 39.5 on the Noometry Index. Claude Haiku 4.5 costs 13× less per token, which makes it the better buy when o1's lead doesn't matter for your workload.
Which is cheaper, Claude Haiku 4.5 or o1?
Claude Haiku 4.5 is cheaper. It lists at $1 per million input tokens and $5 per million output tokens; o1 lists at $15 and $60.
Is Claude Haiku 4.5 or o1 better for coding?
o1 scores higher on coding benchmarks: 46.1 versus 44.0 in the Noometry coding category.
Which has the bigger context window?
Both accept 200K tokens.
How many benchmarks do Claude Haiku 4.5 and o1 share?
28 benchmarks have published results for both models. Claude Haiku 4.5 has 53 scored results on Noometry and o1 has 52.