Claude Sonnet 5.5
| Released | 2026-09-28 |
|---|---|
| Price per 1M | $2.00 in · $10.00 out |
| Overall score | 63.4 · rank 11 |
| Context window | 1,000,000 tokens |
| Provider | Anthropic |
| Weights | Proprietary |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 63.4 | 11 | 50 |
| Overall: Coding | 65.1 | 6 | 41 |
| Overall: Reasoning | 64.2 | 11 | 20 |
| Overall: Agentic work | 64.5 | 7 | 38 |
| Overall: Frontend | 64.7 | 5 | 15 |
| Coding | 65.3 | 6 | 11 |
| Agents | 63.9 | 5 | 14 |
| Reasoning | 65.0 | 12 | 13 |
| Math | 65.7 | 8 | 4 |
| Knowledge | 59.8 | 24 | 3 |
| Frontend | 65.7 | 2 | 3 |
| General | 58.7 | 37 | 1 |
| Multimodal | 57.3 | 34 | 1 |
| Value vs price | +11.5 | 23 |
Buzz
| Posts on X | About 176 a day |
|---|---|
| Mood | 44% positive, 0% negative |
| Hacker News | 31 mentions this week |
| Checked | 2026-11-11 |
- Fable 5.2 output costs $40 per million tokens. Sonnet 5.5 output costs $10. That gap is the whole plan: pay the expensive model for the thinking, pay the cheap one for the typing. so look at who's typing in your setup. if the Fable session writes the code, you're paying $40 f…
- stop using the biggest ai model for everything big models like opus5.5 are smart, but they use up your limits fast. you don't need that much power to read an email or set a reminder for you yesterday i used claude haiku 5.5 medium for all my small tasks and barely used any of …
- THE 10 BEST CODING IDES IN THE AGE OF CODEX AND CLAUDE CODE A coding IDE is no longer only a place you type. OpenAI Codex and Anthropic's Claude Code run as terminal agents that read a repo, edit files, run tests and open pull requests. The editor's job is to host those agents…
Configurations
| Configuration | Balanced | Benchmarks |
|---|---|---|
| Unspecified setting | 63.5 ±3.5 | 9 |
| High | 62.7 ±2.5 | 9 |
| Low | 57.2 ±4.0 | 6 |
| Max | 64.6 ±1.2 | 31 |
| Medium | 58.3 ±2.7 | 7 |
| Extra High | 62.9 ±2.1 | 12 |
Evidence
Coding 65.3 ±1.3
| Benchmark | Result | Source | Date |
|---|---|---|---|
| CursorBench | 55.5% | Epoch AI | 2026-11-11 |
| FrontierSWE | 61.9% | Epoch AI | 2026-11-11 |
| FrontierCode | 52.1% | Epoch AI | 2026-11-11 |
| Code Migration v1.1 | 86.9% | Vals AI | 2026-11-09 |
| LiveBench 2.0 Coding | 77.2% | LiveBench | 2026-11-06 |
| WeirdML v3 | 19.9% | Epoch AI | 2026-11-11 |
| MirrorCode | 97.8% | Epoch AI | 2026-10-05 |
| IOI | 83.1% | Vals AI | 2026-10-29 |
| SWE-bench Multimodal | 58.3% | Vals AI | 2026-11-09 |
| SciCode | 61.0% | Epoch AI | 2026-11-11 |
| ProgramBench | 6.5% | Vals AI | 2026-11-09 |
Agents 63.9 ±0.7
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Terminal-Bench 4 | 63.9% | Terminal-Bench | 2026-09-28 |
| SRE Bench | 30.2% | Vals AI | 2026-11-05 |
| Terminal-Bench-Science | 47.1% | Vals AI | 2026-11-09 |
| Terminal-Bench 4.0 (Vals run) | 64.1% | Vals AI | 2026-11-09 |
| APEX-Agents | 75.5% | Epoch AI | 2026-11-11 |
| LiveBench 2.0 Agentic | 63.4% | LiveBench | 2026-11-06 |
| LiveBench 2.0 Browser Use | 89.3% | LiveBench | 2026-11-06 |
| Terminal-Bench 2.1 (Vals run) | 83.1% | Vals AI | 2026-10-01 |
| EMB v1.1 (Excel financial modeling) | 75.9% | Vals AI | 2026-11-09 |
| LiveBench 2.0 Research | 88.0% | LiveBench | 2026-11-06 |
| Tax Agent Bench | 42.0% | Vals AI | 2026-11-09 |
| Finance Agent v2.1 | 67.3% | Vals AI | 2026-11-09 |
| Vending-Bench 2 | $11,165 | Epoch AI | 2026-11-11 |
| HLAB-Vals 1.1 (legal agent) | 10.4% | Vals AI | 2026-11-09 |
| CyberBench v1.1 | 59.6% | Vals AI | 2026-11-05 |
Reasoning 65.0 ±2.0
| Benchmark | Result | Source | Date |
|---|---|---|---|
| ARC-AGI-2 | 83.9% | Epoch AI | 2026-11-11 |
| ARC-AGI-2 (ARC Prize verified) | 83.9% | ARC Prize | 2026-11-11 |
| Extended NYT Connections | 80.5% | Lech Mazur | 2026-11-10 |
| MysteryMechanism | 49.1% | Vals AI | 2026-11-09 |
| GPQA Diamond | 95.6% | Epoch AI | 2026-09-29 |
| CritPt | 31.4% | Epoch AI | 2026-11-11 |
| ARC-AGI-1 (ARC Prize verified) | 96.5% | ARC Prize | 2026-11-11 |
| ARC-AGI-1 | 96.5% | Epoch AI | 2026-11-11 |
| SimpleBench | 75.9% | SimpleBench | 2026-10-01 |
| BioMysteryBench | 81.1% | Vals AI | 2026-11-09 |
| ARC-AGI-3 | 72.4% | ARC Prize | 2026-11-11 |
| Chess Puzzles | 59.0% | Epoch AI | 2026-10-13 |
| LiveBench 2.0 Reasoning | 91.6% | LiveBench | 2026-11-06 |
| Mystery Game Puzzles | 65.0% | Epoch AI | 2026-09-29 |
| LiveBench 2.0 Data Analysis | 59.5% | LiveBench | 2026-11-06 |
Math 65.7 ±2.2
| Benchmark | Result | Source | Date |
|---|---|---|---|
| FrontierMath Tiers 1-3 | 88.8% | Epoch AI | 2026-09-29 |
| FrontierMath Tier 4 | 80.5% | Epoch AI | 2026-09-29 |
| OTIS Mock AIME 2024-2025 | 100.0% | Epoch AI | 2026-09-29 |
| ProofBench v1.1 | 100.0% | Vals AI | 2026-11-09 |
| ProofBench | 100.0% | Epoch AI | 2026-11-11 |
Knowledge 59.8 ±3.8
| Benchmark | Result | Source | Date |
|---|---|---|---|
| SimpleQA Verified | 46.5% | Epoch AI | 2026-09-29 |
| Humanity's Last Exam | 46.1% | Epoch AI | 2026-11-11 |
| Legal Research Bench | 48.1% | Vals AI | 2026-11-09 |
Frontend 65.7 ±2.1
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LiveBench 2.0 Design | 81.3% | LiveBench | 2026-11-06 |
| Vibe Code Bench v1.1 | 92.4% | Vals AI | 2026-11-09 |
| LMArena WebDev | 1,774 | LMArena | 2026-11-09 |
General 58.7 ±4.1
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LMArena Text | 1,472 | LMArena | 2026-11-09 |
Multimodal 57.3 ±4.1
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LMArena Vision | 1,263 | LMArena | 2026-11-09 |
Results not used in the score
| Benchmark | Result | Source | Why not used |
|---|---|---|---|
| APEX-Agents | 44.6% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-1 | 96.5% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-1 | 92.0% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-1 | 70.0% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-1 | 67.0% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-1 (ARC Prize verified) | 96.5% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-1 (ARC Prize verified) | 92.0% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-1 (ARC Prize verified) | 70.0% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-1 (ARC Prize verified) | 67.0% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-2 | 67.4% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-2 | 32.8% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-2 | 24.0% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-2 (ARC Prize verified) | 67.4% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-2 (ARC Prize verified) | 32.8% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-2 (ARC Prize verified) | 24.0% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-3 | 24.3% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-3 | 1.2% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-3 | 35.5% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-3 | 0.4% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-3 | 4.9% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-3 | 0.4% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-3 | 0.6% | ARC Prize | A stronger configuration result was used |
| CritPt | 31.1% | Epoch AI | A stronger configuration result was used |
| CritPt | 24.6% | Epoch AI | A stronger configuration result was used |
| CritPt | 16.9% | Epoch AI | A stronger configuration result was used |
| CritPt | 11.4% | Epoch AI | A stronger configuration result was used |
| CursorBench | 53.1% | Epoch AI | A stronger configuration result was used |
| CursorBench | 47.8% | Epoch AI | A stronger configuration result was used |
| CursorBench | 39.2% | Epoch AI | A stronger configuration result was used |
| CursorBench | 35.8% | Epoch AI | A stronger configuration result was used |
| LMArena WebDev | 1,713 | LMArena | A stronger configuration result was used |
| SciCode | 57.3% | Epoch AI | A stronger configuration result was used |
| SciCode | 53.7% | Epoch AI | A stronger configuration result was used |
| SciCode | 52.9% | Epoch AI | A stronger configuration result was used |
| SciCode | 49.1% | Epoch AI | A stronger configuration result was used |
Names matched to this model
claude-sonnet-5-5