Model comparison
Claude Haiku 5.5 vs GPT-6 Astra
GPT-6 Astra is the stronger model overall, scoring 70.8 to 49.5 on the Noometry Index. Claude Haiku 5.5 costs 100× less per token, which makes it the better buy when GPT-6 Astra's lead doesn't matter for your workload.
Last verified . 9 shared benchmarks.
Summary
- They share 9 benchmarks with published results for both. Claude Haiku 5.5 scores higher in 0 categories and GPT-6 Astra in 5 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-6 Astra leads 85.1 to 35.2.
- The biggest single-benchmark swing is Mystery Game Puzzles: 30% for Claude Haiku 5.5 and 84% for GPT-6 Astra.
- Claude Haiku 5.5 is cheaper at $0.10 / $0.50 per million input/output tokens, against $10 / $50 for GPT-6 Astra.
- GPT-6 Astra accepts more context: 1.05M tokens versus 1M.
Side by side
| Claude Haiku 5.5 | GPT-6 Astra | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 49.5 | 70.8 |
| Released | 2026-10-07 | 2026-09-03 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 1.05M |
| Max output | 128K | 128K |
| Input $ / M tokens | $0.10 | $10 |
| Output $ / M tokens | $0.50 | $50 |
| Results tracked | 9 | 56 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Astra leads
Claude Haiku 5.5: 49.2 (#50), GPT-6 Astra: 73.7 (#2)
| Benchmark | Claude Haiku 5.5 | GPT-6 Astra |
|---|---|---|
| LMArena WebDev | 1587 | 1786 |
| DeepSWE | — | 74.1% |
| FrontierCode | — | 53.3% |
| FrontierSWE | — | 65.5% |
| SciCode | — | 56.5% |
| GSO | — | 79.4% |
| WeirdML | — | 93.6% |
| LMArena Coding | — | 1487 |
| MirrorCode | — | 46.7% |
| ALE-Bench | — | 2,951 |
Agentic & Tool Use Not comparable
Claude Haiku 5.5: —, GPT-6 Astra: 52.9 (#3)
| Benchmark | Claude Haiku 5.5 | GPT-6 Astra |
|---|---|---|
| APEX-Agents | — | 64.7% |
| Remote Labor Index | — | 20.8% |
| BALROG | — | 68.3% |
| GDP.pdf | — | 34.2% |
| Vending-Bench 2 | — | 15,515 |
Reasoning GPT-6 Astra leads
Claude Haiku 5.5: 35.2 (#69), GPT-6 Astra: 85.1 (#1)
| Benchmark | Claude Haiku 5.5 | GPT-6 Astra |
|---|---|---|
| NYT Connections (extended) | 65.7% | 98.1% |
| Mystery Game Puzzles | 30% | 84% |
| ARC-AGI-2 | — | 95% |
| ARC-AGI-1 | — | 98.5% |
| CritPt | — | 31.7% |
| Chess Puzzles | — | 72% |
| EBR-Bench | — | 76.2% |
| LMArena Hard Prompts | — | 1462 |
| DTBench | — | 97.3% |
| LMCA | — | 64.4% |
| Bench to the Future 3 | — | 0.14 |
| Epoch Capabilities Index | — | 166.45 |
Math GPT-6 Astra leads
Claude Haiku 5.5: 73.6 (#18), GPT-6 Astra: 93.5 (#2)
| Benchmark | Claude Haiku 5.5 | GPT-6 Astra |
|---|---|---|
| FrontierMath (Tiers 1-3) | 75.1% | 93.7% |
| FrontierMath Tier 4 | 46.3% | 97.6% |
| OTIS Mock AIME 2024-2025 | 98.9% | 100% |
| ProofBench | — | 99% |
| LMArena Math | — | 1465 |
| FrontierMath Erdős | — | 2.9% |
Knowledge GPT-6 Astra leads
Claude Haiku 5.5: 50.8 (#70), GPT-6 Astra: 75.3 (#1)
| Benchmark | Claude Haiku 5.5 | GPT-6 Astra |
|---|---|---|
| GPQA Diamond | 89.6% | 95.8% |
| SimpleQA Verified | 23.8% | 75.6% |
| Humanity's Last Exam | — | 54.8% |
| Vectara Hallucination Rate | — | 8.7% |
| LMArena Expert | — | 1483 |
Multimodal GPT-6 Astra leads
Claude Haiku 5.5: 43.7 (#21), GPT-6 Astra: 55.0 (#3)
| Benchmark | Claude Haiku 5.5 | GPT-6 Astra |
|---|---|---|
| Furniture Assembly | 47.5% | 80% |
| LMArena Vision | — | 1281 |
| Blueprint-Bench 2 | — | 49.7% |
| LMArena Document | — | 1468 |
Multilingual Not comparable
Claude Haiku 5.5: —, GPT-6 Astra: 53.7 (#61)
| Benchmark | Claude Haiku 5.5 | GPT-6 Astra |
|---|---|---|
| LMArena Non-English | — | 1430 |
| LMArena Chinese | — | 1484 |
| LMArena French | — | 1456 |
| LMArena German | — | 1440 |
| LMArena Japanese | — | 1379 |
| LMArena Korean | — | 1426 |
| LMArena Russian | — | 1436 |
| LMArena Spanish | — | 1407 |
Instruction Following Not comparable
Claude Haiku 5.5: —, GPT-6 Astra: 76.3 (#44)
| Benchmark | Claude Haiku 5.5 | GPT-6 Astra |
|---|---|---|
| LMArena Instruction Following | — | 1450 |
Long Context Not comparable
Claude Haiku 5.5: —, GPT-6 Astra: 44.5 (#62)
| Benchmark | Claude Haiku 5.5 | GPT-6 Astra |
|---|---|---|
| LMArena Longer Query | — | 1456 |
Writing & Preference Not comparable
Claude Haiku 5.5: —, GPT-6 Astra: 75.3 (#7)
| Benchmark | Claude Haiku 5.5 | GPT-6 Astra |
|---|---|---|
| LMArena Text | — | 1441 |
| LMArena Creative Writing | — | 1418 |
| EQ-Bench Creative Writing | — | 2173 |
| LMArena Multi-Turn | — | 1448 |
Frequently asked questions
Is Claude Haiku 5.5 better than GPT-6 Astra?
GPT-6 Astra is the stronger model overall, scoring 70.8 to 49.5 on the Noometry Index. Claude Haiku 5.5 costs 100× less per token, which makes it the better buy when GPT-6 Astra's lead doesn't matter for your workload.
Which is cheaper, Claude Haiku 5.5 or GPT-6 Astra?
Claude Haiku 5.5 is cheaper. It lists at $0.10 per million input tokens and $0.50 per million output tokens; GPT-6 Astra lists at $10 and $50.
Is Claude Haiku 5.5 or GPT-6 Astra better for coding?
GPT-6 Astra scores higher on coding benchmarks: 73.7 versus 49.2 in the Noometry coding category.
Which has the bigger context window?
GPT-6 Astra does, with 1.05M tokens against 1M.
How many benchmarks do Claude Haiku 5.5 and GPT-6 Astra share?
9 benchmarks have published results for both models. Claude Haiku 5.5 has 9 scored results on Noometry and GPT-6 Astra has 56.