Claude Sonnet 4.5
Anthropic
51.1#139 of 427 overall
- Input / output
- US$ 3,00 / US$ 15,00 US$ per 1M tokens
- Context
- 1M tokens
- Release
- 09/29/2025 release date
- Weights
- Closed
- Speed
- 0 tok/s first token in 0.00 s
Family history
The 8 versions of this family, from first to newest, by release date. Click a point to open that version.
See the data as a table
| Model | Released | Score | Rank | AA Intel. | Elo Arena |
|---|---|---|---|---|---|
| Claude 3 Sonnet | 03/04/2024 | 14.4 | #425 | — | 1218 |
| Claude 3.5 Sonnet | 10/22/2024 | 19.9 | #403 | — | 1298 |
| Claude 3.7 Sonnet | 02/24/2025 | 32.3 | #334 | — | 1314 |
| Claude 4 Sonnet | 05/22/2025 | 37.9 | #281 | — | 1351 |
| Claude Sonnet 4.5 | 09/29/2025 | 51.1 | #139 | — | 1439 |
| Claude Sonnet 4.6 | 02/17/2026 | 70.6 | #38 | — | 1458 |
| Claude Sonnet 5 | 06/30/2026 | 70.8 | #37 | — | 1443 |
| Claude Sonnet 5.5 | 09/28/2026 | 87.5 | #6 | 54.2 | 1473 |
Scores by source
Scores by sourcepercentile = % of ranked models it beats| Metric | Value | Percentile | Source |
|---|---|---|---|
| Arena text (Elo) weight 2 | 1439 | 70% | LMArena |
| Arena WebDev (Elo) weight 1 | 1394 | 29% | LMArena |
| Humanity's Last Exam weight 1 | 17.8 | 66% | Artificial Analysis |
| Terminal-Bench weight 1 | 0.0 | 22% | Artificial Analysis |
| GPQA Diamond | 83.4 | 73% | Artificial Analysis |
| Long-context reasoning | 72.3 | 67% | Artificial Analysis |
By category
| Category | Score | Rank | Metrics |
|---|---|---|---|
| Agentic | 36.0 | #88 | Terminal-Bench |
| Coding | 39.6 | #82 | Arena WebDev (Elo) |
| Reasoning | 63.1 | #124 | Humanity's Last Exam, GPQA Diamond |
| Research | 58.6 | #108 | Long-context reasoning |
| Design | 39.6 | #82 | Arena WebDev (Elo) |
| Chat | 59.9 | #61 | Arena text (Elo) |
History
RankingIA score on each collection day (hover to see the rank).
Daily collection started on 10/30/2026. The chart appears from the second day the model is ranked.