AN
Compare- Context window
- 1M
- Input
- Document · Image · Text
- Output
- Text
- Released
- 2026-06-30
- Max output
- 128K
- Knowledge cutoff
- 2026-01-31
- Weights
- Closed
- Capabilities
- Attachments, Reasoning, Tool call
IN
$2.00
USD / 1,000,000 token
Cache write
$2.50 – $4.00
USD / 1,000,000 token
$2.50 · cache TTL: 5m
$4.00 · cache TTL: 1h
Cache read
$0.20
USD / 1,000,000 token
OUT
$10.00
USD / 1,000,000 token
Performance
Overall capability
54.3/100
- Coding
- 49.1
- Reasoning
- 55.5
- Professional work
- 58.0
- Knowledge & language
- 53.1
| Arena Agent Overall | Professional work | LMArena | 2026-11-08 | |
| MLCR-AA(Artificial Analysis)Claude Sonnet 5 (Low) | Knowledge & language | Artificial Analysis | 2026-11-05 | |
| Humanity’s Last ExamClaude Sonnet 5 (Max) | Reasoning | Artificial Analysis | 2026-11-05 | |
| APEX-Agents 1.1 | Professional work | Mercor | 2026-11-05 | |
| MedCode | Professional work | Vals AI | 2026-11-05 | |
| Mystery Game Puzzles | Reasoning | Epoch AI | 2026-11-05 | |
| Code Migration | Coding | Vals AI | 2026-11-05 | |
| FrontierMath v2 Tiers 1–3 | Reasoning | Epoch AI | 2026-11-05 |
Providers
| Cache write | |||||||
|---|---|---|---|---|---|---|---|
| Anthropic | claude-sonnet-5 | Anthropic messages | global | ||||
| OpenRouter | anthropic/claude-sonnet-5 | Openai chat | global |
ModelRegistry · Catalog updated 2026-11-10