Claude 3.5-Sonnet-20241022: benchmark scores, price and ranking
Claude 3.5-Sonnet-20241022 from Anthropic ranks #22 of 743 LLMs on the LLMs Tiger aggregate benchmark score (54.4), based on 6 benchmark results captured up to 2026-10-13.
- Organisation
- Anthropic
- Type
- Frontier model
- Open weights
- No
- Released
- 2024-10-22 (source: name)
- Cheapest API price
- $3.00 in / $15.00 out per million tokens (via vercel_ai_gateway, observed 2026-10-13T20:47Z)
- Aggregate rank
- #22 of 743 (score 54.4)
Relative to other models, Claude 3.5-Sonnet-20241022 is strongest on Aider Code Editing (#1 of 92) and weakest on SWE-bench Verified (#43 of 73).
Across the 6 benchmarks where at least ten models are scored, Claude 3.5-Sonnet-20241022 is in the top 10% on 2 and in the top half on 5.
Its cheapest listed input price, $3.00 per million tokens, is lower than that of 11% of the 427 models with a listed price.
Claude 3.5-Sonnet-20241022 benchmark results
| Benchmark | Score | Rank | Source | Captured |
|---|---|---|---|---|
| Aider Code Editing | 84.2% | #1 of 92 | Aider Aider Code Editing leaderboard | 2026-10-13 |
| Aider Refactoring | 92.1% | #1 of 12 | Aider Aider Refactoring leaderboard | 2026-10-13 |
| SWE-bench Test | 24.9% | #3 of 11 | SWE-bench Test leaderboard | 2026-10-13 |
| SWE-bench Lite | 30.0% | #8 of 20 | SWE-bench Lite leaderboard | 2026-10-13 |
| Aider Polyglot | 51.6% | #21 of 49 | Aider polyglot leaderboard | 2026-10-13 |
| SWE-bench Verified | 51.6% | #43 of 73 | SWE-bench Verified leaderboard | 2026-10-13 |
About Claude 3.5-Sonnet-20241022
Claude 3.5-Sonnet-20241022 is a frontier model from Anthropic. It was released on 2024-10-22, according to the model's own name. The cheapest API listing we track charges $3.00 per million input tokens and $15.00 per million output tokens, through vercel_ai_gateway. We hold 6 benchmark results for it across 2 areas, and 2 are in the top tenth of their boards.
Where Claude 3.5-Sonnet-20241022 stands
On our aggregate score, Claude 3.5-Sonnet-20241022 is #22 of 743 ranked models, in the top fifty. Its score is 54.4; the scale is centred on 50, so it sits 4.4 points above the typical model. That beats 97% of the ranked field. Among the 70 ranked Anthropic models it is #9. The next one up is Claude 4.5 Opus at 54.5. The next one down is Claude Opus 4.7 (max) at 53.7.
Claude 3.5-Sonnet-20241022, benchmark by benchmark
Aider Code Editing is where Claude 3.5-Sonnet-20241022 does best: 84.2%, first of 92. The next result is gemini-exp-1206 at 80.5%, 3.7 points behind. It covers Aider official benchmark. The board median is 55.6%, so Claude 3.5-Sonnet-20241022 is 28.6 points above it. Source: Aider Aider Code Editing leaderboard, captured 2026-10-13.
Aider Refactoring is where Claude 3.5-Sonnet-20241022 does best: 92.1%, first of 12. The next result is o1-preview at 75.3%, 16.8 points behind. It covers Aider official benchmark. The board median is 50.0%, so Claude 3.5-Sonnet-20241022 is 42.1 points above it. Source: Aider Aider Refactoring leaderboard, captured 2026-10-13.
Claude 3.5-Sonnet-20241022 manages 24.9% on SWE-bench Test, good for #3 of 11. Claude 4.5 Opus leads at 52.6%, 27.7 points further up, and 73% of models sit below. The results either side are CodeAct v2.1 (claude-3-5-sonnet-20241022) (29.4%) and Claude 3.5 Sonnet (18.1%). It covers real GitHub issues resolved. The board median is 12.5%, so Claude 3.5-Sonnet-20241022 is 12.4 points above it. Source: SWE-bench Test leaderboard, captured 2026-10-13.
SWE-bench Lite puts Claude 3.5-Sonnet-20241022 at 30.0%, #8 of 20. That beats 60% of the models tested, but 26.7 points short of Claude 4 Sonnet (56.7%). The results either side are o3-mini_1.0 (30.3%) and CodeAct v1.8 (26.7%). It covers real GitHub issues resolved. The board median is 24.7%, so Claude 3.5-Sonnet-20241022 is 5.3 points above it. Source: SWE-bench Lite leaderboard, captured 2026-10-13.
On Aider Polyglot, Claude 3.5-Sonnet-20241022 reaches 51.6%. It ranks #21 of 49, ahead of 57% of the field, and 31.5 points behind gemini-2.5-pro-preview-06-05 at 83.1%. The results either side are GPT-4.1 (52.4%) and DeepSeek Chat V3 (prev) (48.4%). It covers edits across 6 languages (225 exercises). The board median is 44.9%, so Claude 3.5-Sonnet-20241022 is 6.7 points above it. Source: Aider polyglot leaderboard, captured 2026-10-13.
Claude 3.5-Sonnet-20241022 lands below the middle on SWE-bench Verified: 51.6%, #43 of 73, with 58% of the field scoring higher. Claude Opus 4.7 (max) leads at 83.5%; the closest result above is Frogboss 32B 2510 at 53.6%. It covers real GitHub issues resolved. The board median is 56.4%, so Claude 3.5-Sonnet-20241022 is 4.8 points below it. Source: SWE-bench Verified leaderboard, captured 2026-10-13.
Strengths and weaknesses
Software engineering: solid, ahead of 64% of the field on average across 4 benchmarks. Its best board here is Aider Code Editing (#1 of 92) and its weakest is SWE-bench Verified (#43 of 73). General knowledge and chat: strong, ahead of 82% of the field on average across 2 benchmarks. Its best board here is Aider Refactoring (#1 of 12) and its weakest is SWE-bench Test (#3 of 11).
Price and value
The median input price across the 427 priced models is $0.500 per million tokens, so Claude 3.5-Sonnet-20241022 costs more than the typical listing. Models that score within 1.5 points or better and cost less: DeepSeek V4 Flash 0731 (max) ($0.016, 53.1), GLM-5.3 (max) ($0.079, 53.8), GPT-6 Luna (max) ($0.100, 53.7). Similar scores cost more on Claude Opus 4.7 (max) ($5.00) and Claude Mythos Preview (Early) ($10.00).
Release history
The Anthropic model released before it was Claude 3.5 Haiku (Oct 2024) on 2024-10-22, which scores 41.3. The next one was Claude 3.5 Sonnet (Oct 2024) on 2024-10-22, at 45.4.
Reading these numbers
The 6 scores on this page come from 6 sources: Aider Aider Code Editing leaderboard (1), Aider polyglot leaderboard (1), Aider Aider Refactoring leaderboard (1), SWE-bench Lite leaderboard (1). 4 of them come from an official leaderboard and 2 from datasets or third-party runs. With 6 results the aggregate rank can move a lot when one new score arrives.
Models ranked nearby
- #19 mimo-v2.6-pro (54.7)
- #20 Claude 4 Sonnet (54.5)
- #21 Claude 4.5 Opus (54.5)
- #23 DeepSeek V4 Pro 0813 (max) (54.3)
- #24 GPT-5.3 Codex (xhigh) (54.3)
- #25 Qwen 3.6 Max (Preview) (54.1)
More from Anthropic
- Claude Opus 5.5 (max) (#1, 64.3)
- Claude Fable 5.1 (max) (#3, 61.9)
- Claude Sonnet 5.5 (max) (#4, 60.8)
- Claude Opus 5 (max) (#8, 60.0)
- Claude Fable 5 (max) (#9, 59.9)
- claude-opus-4-8-xhigh-effort (#14, 56.6)
Compare it on the live leaderboard → · How the aggregate score works