Model comparison
Claude Haiku 4.5 vs GPT-6 Sol
GPT-6 Sol is the stronger model overall, scoring 61.8 to 39.5 on the Noometry Index. Claude Haiku 4.5 costs 2.0× less per token, which makes it the better buy when GPT-6 Sol's lead doesn't matter for your workload.
Last verified . 33 shared benchmarks.
Summary
- They share 33 benchmarks with published results for both. Claude Haiku 4.5 scores higher in 1 category and GPT-6 Sol in 9 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-6 Sol leads 74.0 to 15.1.
- The biggest single-benchmark swing is ARC-AGI-2: 4% for Claude Haiku 4.5 and 89.6% for GPT-6 Sol.
- Claude Haiku 4.5 is cheaper at $1 / $5 per million input/output tokens, against $2 / $10 for GPT-6 Sol.
- GPT-6 Sol accepts more context: 1.05M tokens versus 200K.
Side by side
| Claude Haiku 4.5 | GPT-6 Sol | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 39.5 | 61.8 |
| Released | 2025-10-15 | 2026-09-22 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | 1.05M |
| Max output | 64K | 128K |
| Input $ / M tokens | $1 | $2 |
| Output $ / M tokens | $5 | $10 |
| Results tracked | 53 | 45 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Sol leads
Claude Haiku 4.5: 44.0 (#78), GPT-6 Sol: 60.1 (#11)
| Benchmark | Claude Haiku 4.5 | GPT-6 Sol |
|---|---|---|
| LMArena WebDev | 1330 | 1688 |
| SciCode | 43.3% | 57.6% |
| LMArena Coding | 1453 | 1447 |
| ALE-Bench | 653.48 | 2,462 |
| DeepSWE | — | 68.8% |
| FrontierCode | — | 49.3% |
| SWE-bench Verified (bash only) | 66.6% | — |
| SWE-bench Multilingual | 64.7% | — |
| WeirdML | 45.4% | — |
Agentic & Tool Use GPT-6 Sol leads
Claude Haiku 4.5: 33.6 (#52), GPT-6 Sol: 37.2 (#36)
| Benchmark | Claude Haiku 4.5 | GPT-6 Sol |
|---|---|---|
| Vending-Bench 2 | 458.89 | 14,428 |
| Terminal-Bench | 35.5% | — |
| APEX-Agents | — | 54.3% |
| Berkeley Function Calling Leaderboard | 68.7% | — |
| DeepResearch Bench | 45.5% | — |
| BALROG | 31.2% | — |
| ExploitBench | 13.7% | — |
| GDP.pdf | — | 26.4% |
Reasoning GPT-6 Sol leads
Claude Haiku 4.5: 15.1 (#320), GPT-6 Sol: 74.0 (#9)
| Benchmark | Claude Haiku 4.5 | GPT-6 Sol |
|---|---|---|
| ARC-AGI-2 | 4% | 89.6% |
| NYT Connections (extended) | 14.3% | 90.1% |
| ARC-AGI-1 | 47.7% | 95.5% |
| CritPt | 0% | 30.9% |
| LMArena Hard Prompts | 1420 | 1418 |
| DTBench | 73.6% | 97.3% |
| LMCA | 30.9% | 59.1% |
| Epoch Capabilities Index | 142.41 | 162.72 |
| Chess Puzzles | 8% | — |
| EBR-Bench | — | 53.3% |
| Mystery Game Puzzles | — | 56% |
| ForecastBench | 61.4 | — |
Math GPT-6 Sol leads
Claude Haiku 4.5: 44.9 (#78), GPT-6 Sol: 87.2 (#7)
| Benchmark | Claude Haiku 4.5 | GPT-6 Sol |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 66.7% | 100% |
| LMArena Math | 1396 | 1402 |
| FrontierMath (Tiers 1-3) | — | 89.8% |
| FrontierMath Tier 4 | — | 90% |
| ProofBench | — | 83% |
| Omni-MATH | 56.1% | — |
| MATH Level 5 | 96.4% | — |
| FrontierMath (Feb 2025 set) | 5.9% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge GPT-6 Sol leads
Claude Haiku 4.5: 37.7 (#153), GPT-6 Sol: 64.8 (#15)
| Benchmark | Claude Haiku 4.5 | GPT-6 Sol |
|---|---|---|
| GPQA Diamond | 71.2% | 94.3% |
| SimpleQA Verified | 13.2% | 60.7% |
| Vectara Hallucination Rate | 9.8% | 6.5% |
| LMArena Expert | 1442 | 1439 |
| MMLU-Pro | 77.7% | — |
| GPQA (HELM) | 60.5% | — |
Multimodal GPT-6 Sol leads
Claude Haiku 4.5: 26.8 (#118), GPT-6 Sol: 47.6 (#10)
| Benchmark | Claude Haiku 4.5 | GPT-6 Sol |
|---|---|---|
| Blueprint-Bench 2 | 0% | 36.9% |
| LMArena Vision | — | 1245 |
| Furniture Assembly | — | 58.3% |
| LMArena Document | 1420 | — |
Multilingual Too close to call
Claude Haiku 4.5: 49.9 (#129), GPT-6 Sol: 50.5 (#118)
| Benchmark | Claude Haiku 4.5 | GPT-6 Sol |
|---|---|---|
| LMArena Non-English | 1377 | 1385 |
| LMArena Chinese | 1417 | 1405 |
| LMArena French | 1408 | 1410 |
| LMArena German | 1375 | 1390 |
| LMArena Japanese | 1339 | 1385 |
| LMArena Korean | 1347 | 1341 |
| LMArena Russian | 1381 | 1401 |
| LMArena Spanish | 1420 | 1384 |
Instruction Following GPT-6 Sol leads
Claude Haiku 4.5: 71.4 (#149), GPT-6 Sol: 74.5 (#94)
| Benchmark | Claude Haiku 4.5 | GPT-6 Sol |
|---|---|---|
| LMArena Instruction Following | 1414 | 1412 |
| IFEval | 80.1% | — |
Long Context Too close to call
Claude Haiku 4.5: 43.6 (#92), GPT-6 Sol: 43.1 (#108)
| Benchmark | Claude Haiku 4.5 | GPT-6 Sol |
|---|---|---|
| LMArena Longer Query | 1427 | 1411 |
Writing & Preference GPT-6 Sol leads
Claude Haiku 4.5: 57.9 (#123), GPT-6 Sol: 71.9 (#18)
| Benchmark | Claude Haiku 4.5 | GPT-6 Sol |
|---|---|---|
| LMArena Text | 1396 | 1395 |
| LMArena Creative Writing | 1372 | 1378 |
| LMArena Multi-Turn | 1409 | 1412 |
| EQ-Bench Creative Writing | — | 2125 |
| WildBench | 83.9% | — |
| EQ-Bench 4 | 1064 | — |
Frequently asked questions
Is Claude Haiku 4.5 better than GPT-6 Sol?
GPT-6 Sol is the stronger model overall, scoring 61.8 to 39.5 on the Noometry Index. Claude Haiku 4.5 costs 2.0× less per token, which makes it the better buy when GPT-6 Sol's lead doesn't matter for your workload.
Which is cheaper, Claude Haiku 4.5 or GPT-6 Sol?
Claude Haiku 4.5 is cheaper. It lists at $1 per million input tokens and $5 per million output tokens; GPT-6 Sol lists at $2 and $10.
Is Claude Haiku 4.5 or GPT-6 Sol better for coding?
GPT-6 Sol scores higher on coding benchmarks: 60.1 versus 44.0 in the Noometry coding category.
Which has the bigger context window?
GPT-6 Sol does, with 1.05M tokens against 200K.
How many benchmarks do Claude Haiku 4.5 and GPT-6 Sol share?
33 benchmarks have published results for both models. Claude Haiku 4.5 has 53 scored results on Noometry and GPT-6 Sol has 45.