Model vendor
Anthropic
AI safety company
Updated Aug 10, 2026. Data as of 2026-08-07
Published models20
Published versions44
Open defaults0
CountryUS
Anthropic models
Default-version specifications and scored-version capability values are labeled separately.
| AN | Claude Opus 4.8 | 82.9 | 80.0% | 1M | Closed weights | Claude Opus 4.8 | |
| AN | Claude Opus 4.7 | 78.4 | 80.0% | 1M | Closed weights | Claude Opus 4.7 | |
| AN | Claude 3.5 Sonnet | 73.4 | 80.0% | 200K | Closed weights | Claude 3.5 Sonnet | |
| AN | Claude Opus 4.6 | 69.2 | 80.0% | 1M | Closed weights | Claude Opus 4.6 | |
| AN | Claude 3.7 Sonnet | 63.6 | 60.0% | 200K | Closed weights | Claude 3.7 Sonnet | |
| AN | Claude 3 Opus | 58.8 | 80.0% | 200K | Closed weights | Claude 3 Opus | |
| AN | Claude 3.5 Haiku | 41.2 | 80.0% | 200K | Closed weights | Claude 3.5 Haiku | |
| AN | Claude 3 Sonnet | 40.0 | 80.0% | 200K | Closed weights | Claude 3 Sonnet | |
| AN | Claude 3 Haiku | 34.1 | 80.0% | 200K | Closed weights | Claude 3 Haiku | |
| AN | Claude Fable 5 | N/A | N/A | 1M | Closed weights | Not scored | |
| AN | Claude Haiku 4.5 | N/A | N/A | 200K | Closed weights | Not scored | |
| AN | Claude Mythos Preview | N/A | N/A | N/A | Closed weights | N/A | Not scored |
| AN | Claude Opus 4 | N/A | N/A | 200K | Closed weights | Not scored | |
| AN | Claude Opus 4.1 | N/A | N/A | 200K | Closed weights | Not scored | |
| AN | Claude Opus 4.5 | N/A | N/A | 200K | Closed weights | Not scored | |
| AN | Claude Opus 5 | N/A | N/A | 1M | Closed weights | Not scored | |
| AN | Claude Sonnet 4 | N/A | N/A | 200K | Closed weights | Not scored | |
| AN | Claude Sonnet 4.5 | N/A | N/A | 200K | Closed weights | Not scored | |
| AN | Claude Sonnet 4.6 | N/A | N/A | 1M | Closed weights | Not scored | |
| AN | Claude Sonnet 5 | N/A | N/A | 1M | Closed weights | Not scored |
Anthropic capability scores
Each unique model uses its latest scored version.
Anthropic highlights
Leaders from this vendor's current published catalog.
Top scoreClaude Opus 4.882.9 LLMBoardLowest official inputClaude Opus 4.8$5.0 / 1MLargest contextClaude Fable 51M tokens
Open-weight defaults0of 20 published models
What is Anthropic?
Vendor context from the published model registry.
AI safety company
Model ownership is separate from API hosting. One model can be offered by multiple API providers with different prices.
- Vendor
- Anthropic
- Published models
- 20
- Published versions
- 44
- Country
- US
- Founded
- Not published
Vendor description and inventory come from the published model registry.
FAQ
Common questions about Anthropic.
What is Anthropic's highest-ranked model?
Claude Opus 4.8 currently has the strongest published LLMBoard score from this vendor.
How many Anthropic models are tracked?
The current public catalog contains 20 independently selectable Anthropic model products and 44 published versions.
Does Anthropic host every listed model API?
Not necessarily. Vendor identifies the model developer; API providers are tracked separately.
Does Anthropic publish open-weight models?
No current default version is marked as open weight.