Model Tracker
Claude Sonnet 4.5
Claude Sonnet 4.5 is an AI model made by Anthropic. Anthropic released it on 29 September 2025. It is deprecated. It takes text, image and returns text.
In inglese
Claude Sonnet 4.5 is an AI model made by Anthropic. Anthropic released it on 29 September 2025. It is deprecated. It takes text, image and returns text. Its context window is 200,000 tokens. Its longest answer is 64,000 tokens. Through the API it costs $3.00 per 1M tokens for input and $15.00 per 1M tokens for output. Its knowledge cutoff is January 2025.
- Released
- 29 September 2025 — platform.claude.com, read 16 October 2026
- Status
- Deprecated — platform.claude.com, read 16 October 2026
- Takes
- Text, Image — platform.claude.com, read 16 October 2026
- Returns
- Text — platform.claude.com, read 16 October 2026
- Open weights
- Not confirmed: The maker's cited model and product pages do not state whether the model's weights are open or downloadable.
- Licence
- Not confirmed: The maker's cited model pages do not name a licence for the model weights.
- Context window
- 200,000 tokens — platform.claude.com, read 16 October 2026
- Longest answer
- 64,000 tokens — platform.claude.com, read 16 October 2026
- Knowledge cutoff
- January 2025 — platform.claude.com, read 16 October 2026
- API price, input
- $3.00 per 1M tokens — platform.claude.com, read 16 October 2026
- API price, output
- $15.00 per 1M tokens — platform.claude.com, read 16 October 2026
- Where
- Not confirmed: two readings disagree: api, cloud and api, app, cloud
- Parameters
- Not confirmed: The maker's cited pages do not publish a parameter count for this model.
- Model card
- Not confirmed: two readings disagree: https://assets.anthropic.com/m/12f214efcc2f457a/original/Claude-Sonnet-4-5-System-Card.pdf and https://www.anthropic.com/claude-sonnet-4-5-system-card
Benchmarks
Measured by others, not by the maker: each score with the name the measurer gives the model and the setting it ran, its day and its licence.
Arena
- Text Arena, claude-sonnet-4-5-20250929-high-32k: 1,456 (1,454–1,459, 83,940 votes), published 15 October 2026
- Text Arena, claude-sonnet-4-5-20250929: 1,455 (1,452–1,458, 82,688 votes), published 15 October 2026
Arena’s leaderboard dataset (lmarena-ai/leaderboard-dataset on Hugging Face), CC BY 4.0: ratings from people’s votes between two anonymous models, as Arena publishes them. Data · CC BY 4.0
Epoch AI
- Epoch Capabilities Index, Claude Sonnet 4.5: 146.8 (145.2–148.4), as of 19 October 2026
- GPQA Diamond, claude-sonnet-4-5-20250929_59K: 82.3% (± 2.7), run 28 October 2025
- GPQA Diamond, claude-sonnet-4-5-20250929_32K: 81.7% (± 2.8), run 21 October 2025
- GPQA Diamond, claude-sonnet-4-5-20250929_16K: 78.8% (± 2.9), run 28 October 2025
- GPQA Diamond, claude-sonnet-4-5-20250929: 73.7% (± 3.1), run 29 September 2025
- FrontierMath, tiers 1–3, claude-sonnet-4-5-20250929_32K: 23.9% (± 2.5), run 11 June 2026
- FrontierMath, tier 4, claude-sonnet-4-5-20250929_32K: 2.4% (± 2.4), run 11 June 2026
- SWE-bench Verified, claude-sonnet-4-5-20250929: 71.3% (± 2.1), run 5 February 2026
- OTIS Mock AIME 2024–2025, claude-sonnet-4-5-20250929_59K: 77.8% (± 6.3), run 28 October 2025
- OTIS Mock AIME 2024–2025, claude-sonnet-4-5-20250929_32K: 77.8% (± 6.3), run 21 October 2025
- OTIS Mock AIME 2024–2025, claude-sonnet-4-5-20250929_16K: 71.1% (± 6.8), run 28 October 2025
- OTIS Mock AIME 2024–2025, claude-sonnet-4-5-20250929: 35.6% (± 7.2), run 29 September 2025
- SimpleQA Verified, claude-sonnet-4-5-20250929_59K: 30.7% (± 1.5), run 27 August 2026
- SimpleQA Verified, claude-sonnet-4-5-20250929: 23.7% (± 1.3), run 10 August 2026
Epoch AI, ‘Capabilities & benchmarking’, epoch.ai/benchmarks, CC BY 4.0: Epoch AI’s own runs and its Capabilities Index. Data · CC BY 4.0
Questions
When was Claude Sonnet 4.5 released?
Anthropic released Claude Sonnet 4.5 on 29 September 2025.
Is Claude Sonnet 4.5 still available?
It is deprecated: Anthropic has announced the date it will retire it.
What is the context window of Claude Sonnet 4.5?
The context window of Claude Sonnet 4.5 is 200,000 tokens, as Anthropic prints it.
How much does the Claude Sonnet 4.5 API cost?
On the standard tier, Claude Sonnet 4.5 costs $3.00 per 1M tokens for input and $15.00 per 1M tokens for output, as Anthropic lists it.
What does Claude Sonnet 4.5 take and return?
Claude Sonnet 4.5 takes text, image and returns text.
What is the knowledge cutoff of Claude Sonnet 4.5?
The knowledge cutoff of Claude Sonnet 4.5 is January 2025.
More from Anthropic
- Claude 1.0
- Claude 1.1
- Claude 1.2
- Claude 1.3
- Claude 2
- Claude 2.1
- Claude 3 Haiku
- Claude 3 Opus
- Claude 3 Sonnet
- Claude 3.5 Haiku
- Claude Fable 5
- Claude Fable 5.1
- Claude Haiku 4.5
- Claude Haiku 5.5
- Claude Instant 1.0
- Claude Instant 1.1
- Claude Instant 1.2
- Claude Mythos Preview
- Claude Opus 4.1
- Claude Opus 4.5
- Claude Opus 4.6
- Claude Opus 4.7
- Claude Opus 4.8
- Claude Opus 5
Sources
- Claude Sonnet 4.5 - Claude Platform Docs (read 16 October 2026)
- Model deprecations - Claude Platform Docs (read 16 October 2026)
- Pricing - Claude Platform Docs (read 16 October 2026)
- Anthropic’s Transparency Hub (read 16 October 2026)
- Claude Sonnet 4.5 System Card (read 16 October 2026)