Claude Sonnet 5.5 benchmarks
Claude Sonnet 5.5 is a proprietary model from Anthropic, released 28 Sep 2026. It ranks 1st of 54 on our overall leaderboard and 2nd of 54 for reasoning, though some of its results are not yet published. At A$5.76 per million tokens (blended), it costs 1.3× the median model we track. It writes about 145 tokens a second, and answers start after 4.6 min on average, thinking included. It can be called from AWS Bedrock in Sydney and Melbourne and Google Vertex AI in Sydney, but only through global routing, so requests may be processed outside Australia.
- CONTEXT WINDOW
- 1,000,000 tokens
- MAX OUTPUT
- 128,000 tokens
- INPUT
- text, image
- WEIGHTS
- Proprietary
- RELEASED
- 28 Sep 2026
- OUTPUT SPEED
- 145 tok/s
- ANSWER STARTS AFTER
- 4.6 min (thinking included)
Claude Sonnet 5.5 across our rankings.
Benchmark results.
Independent test scores are for the adaptive reasoning, max effort setting. “Among tracked” ranks Claude Sonnet 5.5 against the 54 models on this site; LMArena ranks run across its full leaderboard.
| BENCHMARK | RESULT | AMONG TRACKED | SOURCE DETAIL |
|---|---|---|---|
| INDEPENDENT TESTS · ARTIFICIAL ANALYSIS | |||
| AA Intelligence Index | 56.0 | #2 of 54 | Adaptive reasoning, max effort |
| Humanity’s Last Exam | 55.0% | #5 of 54 | Adaptive reasoning, max effort |
| AA-LCR | 82.7% | #16 of 54 | Adaptive reasoning, max effort |
| SciCode | 61.0% | #4 of 49 | Adaptive reasoning, max effort |
| BLIND HUMAN VOTES · LMARENA | |||
| LMArena WebDev1,810 votes · ±16 | 1709 | #5 of 52 | #5 of 137 on LMArena · 30 Sep 2026 |
What Claude Sonnet 5.5 costs in Australian dollars.
List prices converted at the snapshot’s RBA rate, excluding GST. Reasoning models are billed for their thinking as output tokens, so heavier settings cost more than the job estimates show.
| PER MILLION TOKENS | AUD | USD LIST |
|---|---|---|
| Input tokens | A$2.88 | US$2.00 |
| Output tokens | A$14.39 | US$10.00 |
| Blended (3 in : 1 out) | A$5.76 | US$4.00 |
| JOB | COST (AUD) | MEDIAN MODEL |
|---|---|---|
| Customer support replyPER 1,000 REPLIES | A$10.07 | ≈ A$8.20 |
| Summarise a 30-page documentPER 100 DOCUMENTS | A$6.91 | ≈ A$5.35 |
| Agentic coding taskPER 10 TASKS | A$5.47 | ≈ A$4.27 |
| SETTING | INTELLIGENCE | A$ / 1M | SPEED |
|---|---|---|---|
| Adaptive reasoning, max effort | 56.0 | A$5.76 | 145 tok/s |
| Adaptive reasoning, xhigh effort | 51.9 | A$5.76 | 109 tok/s |
| Adaptive reasoning, high effort | 46.7 | A$5.76 | 108 tok/s |
| Adaptive reasoning, medium effort | 40.7 | A$5.76 | 109 tok/s |
| Adaptive reasoning, low effort | A$5.76 | 104 tok/s |
Running Claude Sonnet 5.5 in Australia.
It can be called from AWS Bedrock in Sydney and Melbourne and Google Vertex AI in Sydney, but only through global routing, so requests may be processed outside Australia. How the platforms compare →
| PLATFORM | AVAILABILITY | DETAIL |
|---|---|---|
| AWS Bedrock · Sydney | GLOBAL | |
| AWS Bedrock · Melbourne | GLOBAL | |
| Azure · Australia East | NOT IN AU | |
| Google Vertex AI · Sydney | GLOBAL |
Compare it with.
Claude Sonnet 5.5, answered.
How much does Claude Sonnet 5.5 cost in Australian dollars?
Claude Sonnet 5.5 lists at US$2.00 per million input tokens and US$10.00 per million output tokens — A$2.88 and A$14.39 at A$1 = US$0.6949 (RBA, 1 Oct 2026), excluding GST. A typical customer support reply works out at about A$10.07 per 1,000 replies, before any reasoning tokens.
Can I use Claude Sonnet 5.5 in Australia with data kept onshore?
Claude Sonnet 5.5 can be called from AWS Bedrock in Sydney and Melbourne and Google Vertex AI in Sydney, but only through global routing, so requests may be processed outside Australia. Availability comes from the cloud providers’ own documentation; check your provider’s current terms before relying on it for data-residency obligations.
How fast is Claude Sonnet 5.5?
Artificial Analysis measures Claude Sonnet 5.5 at about 145 tok/s of output, with the answer starting after 4.6 min on average once thinking time is included (adaptive reasoning, max effort setting).
What is Claude Sonnet 5.5’s context window?
Anthropic documents a context window of 1,000,000 tokens (1M), with up to 128,000 output tokens. On AA-LCR, which tests reasoning across ~100,000-token document sets, it scores 82.7%.
Ratings: LMArena leaderboard dataset (CC BY 4.0), rescaled for the lens scores · leaderboards to 30 Sep 2026
Evaluations, prices and speed: Artificial Analysis (artificialanalysis.ai) · fetched 1 Oct 2026
Exchange rate: Reserve Bank of Australia, table F11.1 · A$1 = US$0.6949 on 1 Oct 2026 · prices exclude GST
Snapshot 2 Oct 2026 · updated weekly · How the rankings work →