Model comparison
Claude Haiku 4.5 vs o3-mini
Claude Haiku 4.5 is the stronger model overall, scoring 39.3 to 36.7 on the Noometry Index.
Last verified . 33 shared benchmarks.
Summary
- They share 33 benchmarks with published results for both. Claude Haiku 4.5 scores higher in 6 categories and o3-mini in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Haiku 4.5 leads 43.0 to 27.6.
- The biggest single-benchmark swing is ARC-AGI-1: 47.7% for Claude Haiku 4.5 and 34.5% for o3-mini.
- Both cost about the same: $1 input and $5 output per million tokens.
Side by side
| Claude Haiku 4.5 | o3-mini | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 39.3 | 36.7 |
| Released | 2025-10-15 | 2024-12-20 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | 200K |
| Max output | 64K | 100K |
| Input $ / M tokens | $1 | $1.10 |
| Output $ / M tokens | $5 | $4.40 |
| Results tracked | 53 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Haiku 4.5 leads
Claude Haiku 4.5: 43.9 (#71), o3-mini: 41.4 (#116)
| Benchmark | Claude Haiku 4.5 | o3-mini |
|---|---|---|
| SciCode | 43.3% | 39.8% |
| WeirdML | 45.4% | 43.7% |
| LMArena Coding | 1454 | 1378 |
| SWE-bench Verified (bash only) | 66.6% | — |
| Aider Polyglot | — | 60.4% |
| LMArena WebDev | 1330 | — |
| SWE-bench Multilingual | 64.7% | — |
| GSO | — | 1.3% |
| LiveBench Coding | — | 82.7% |
| CadEval | — | 54% |
| ALE-Bench | 653.48 | — |
Agentic & Tool Use Claude Haiku 4.5 leads
Claude Haiku 4.5: 33.5 (#50), o3-mini: 29.3 (#79)
| Benchmark | Claude Haiku 4.5 | o3-mini |
|---|---|---|
| Terminal-Bench | 35.5% | — |
| Berkeley Function Calling Leaderboard | 68.7% | — |
| Cybench | — | 22.5% |
| DeepResearch Bench | 45.5% | — |
| BALROG | 31.2% | — |
| ExploitBench | 13.7% | — |
| Vending-Bench 2 | 458.89 | — |
Reasoning o3-mini leads
Claude Haiku 4.5: 15.1 (#310), o3-mini: 16.3 (#295)
| Benchmark | Claude Haiku 4.5 | o3-mini |
|---|---|---|
| ARC-AGI-2 | 4% | 3% |
| ARC-AGI-1 | 47.7% | 34.5% |
| CritPt | 0% | 0.3% |
| Chess Puzzles | 8% | 17% |
| LMArena Hard Prompts | 1420 | 1366 |
| DTBench | 73.6% | 68.8% |
| LMCA | 30.9% | 19% |
| Epoch Capabilities Index | 142.42 | 140.35 |
| ForecastBench | 61.4 | 59.6 |
| SimpleBench | — | 22.8% |
| NYT Connections (extended) | 14.3% | — |
| LiveBench Reasoning | — | 89.6% |
| Mystery Game Puzzles | — | 7% |
| LiveBench Data Analysis | — | 70.6% |
| LiveBench | — | 75.9% |
Math Claude Haiku 4.5 leads
Claude Haiku 4.5: 43.0 (#69), o3-mini: 27.6 (#236)
| Benchmark | Claude Haiku 4.5 | o3-mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 66.7% | 76.9% |
| LMArena Math | 1395 | 1396 |
| MATH Level 5 | 96.4% | 96.5% |
| FrontierMath (Feb 2025 set) | 5.9% | 12.4% |
| FrontierMath Tier 4 (v1) | 2.1% | 4.2% |
| FrontierMath (Tiers 1-3) | — | 18.6% |
| FrontierMath Tier 4 | — | 0% |
| Omni-MATH | 56.1% | — |
| LiveBench Math | — | 77.3% |
Knowledge Too close to call
Claude Haiku 4.5: 37.6 (#146), o3-mini: 38.3 (#135)
| Benchmark | Claude Haiku 4.5 | o3-mini |
|---|---|---|
| GPQA Diamond | 71.2% | 77% |
| SimpleQA Verified | 13.2% | 15.3% |
| LMArena Expert | 1443 | 1364 |
| MMLU-Pro | 77.7% | — |
| Confabulations | — | 17.9% |
| Vectara Hallucination Rate | 9.8% | — |
| GPQA (HELM) | 60.5% | — |
Multimodal Not comparable
Claude Haiku 4.5: 27.0 (#111), o3-mini: —
| Benchmark | Claude Haiku 4.5 | o3-mini |
|---|---|---|
| Blueprint-Bench 2 | 0% | — |
| LMArena Document | 1420 | — |
Multilingual Claude Haiku 4.5 leads
Claude Haiku 4.5: 50.1 (#124), o3-mini: 45.9 (#159)
| Benchmark | Claude Haiku 4.5 | o3-mini |
|---|---|---|
| LMArena Non-English | 1376 | 1319 |
| LMArena Chinese | 1415 | 1379 |
| LMArena French | 1408 | 1333 |
| LMArena German | 1377 | 1303 |
| LMArena Japanese | 1339 | 1286 |
| LMArena Korean | 1346 | 1314 |
| LMArena Russian | 1380 | 1304 |
| LMArena Spanish | 1420 | 1321 |
Instruction Following o3-mini leads
Claude Haiku 4.5: 71.6 (#144), o3-mini: 75.3 (#67)
| Benchmark | Claude Haiku 4.5 | o3-mini |
|---|---|---|
| LMArena Instruction Following | 1414 | 1338 |
| LiveBench Instruction Following | — | 84.4% |
| IFEval | 80.1% | — |
Long Context Claude Haiku 4.5 leads
Claude Haiku 4.5: 43.6 (#86), o3-mini: 33.8 (#251)
| Benchmark | Claude Haiku 4.5 | o3-mini |
|---|---|---|
| LMArena Longer Query | 1427 | 1343 |
| Fiction.LiveBench | — | 50% |
Writing & Preference Claude Haiku 4.5 leads
Claude Haiku 4.5: 58.1 (#119), o3-mini: 50.4 (#177)
| Benchmark | Claude Haiku 4.5 | o3-mini |
|---|---|---|
| LMArena Text | 1396 | 1337 |
| LMArena Creative Writing | 1371 | 1286 |
| LMArena Multi-Turn | 1409 | 1320 |
| Short-Story Creative Writing | — | 61.7% |
| WildBench | 83.9% | — |
| EQ-Bench 4 | 1064 | — |
| LiveBench Language | — | 50.7% |
Frequently asked questions
Is Claude Haiku 4.5 better than o3-mini?
Claude Haiku 4.5 is the stronger model overall, scoring 39.3 to 36.7 on the Noometry Index.
Which is cheaper, Claude Haiku 4.5 or o3-mini?
o3-mini is cheaper. It lists at $1.10 per million input tokens and $4.40 per million output tokens; Claude Haiku 4.5 lists at $1 and $5.
Is Claude Haiku 4.5 or o3-mini better for coding?
Claude Haiku 4.5 scores higher on coding benchmarks: 43.9 versus 41.4 in the Noometry coding category.
Which has the bigger context window?
Both accept 200K tokens.
How many benchmarks do Claude Haiku 4.5 and o3-mini share?
33 benchmarks have published results for both models. Claude Haiku 4.5 has 53 scored results on Noometry and o3-mini has 51.