← All models

Claude 3.5 Sonnet

Language Retired Full comparison ↗

anthropic/claude-3-5-sonnet · by Anthropic · decoder-only-transformer

Pricing — 1 offering(s)

Anthropic

Input tokens

Retired — no longer sold · last price $3.00 / 1M tokens (input), until 2025-10-27

  • $3.00 / 1M tokens (input) Historical 2024-10-22 → 2025-10-27

Output tokens

Retired — no longer sold · last price $15.00 / 1M tokens (output), until 2025-10-27

  • $15.00 / 1M tokens (output) Historical 2024-10-22 → 2025-10-27

Showing the active price and any recorded history. Full pricing history is available via the paid API — see API docs.

See how Claude 3.5 Sonnet fits into a cost-aware routing setup

See how →

Capability profile

How Claude 3.5 Sonnet rates across core capability dimensions, with the task-level evidence behind each rating.

reasoning Strong
coding Moderate

SWE-bench Verified (500 human-verified GitHub issues; patch must pass the repo's own tests; no file-location hints): resolves 50.8% — fails roughly 1 issue in 2. Independent (official SWE-bench leaderboard; agentless harness). Best-in-class at its 2024 launch, now far behind the frontier.

tool use Strong
instruction following Strong
context window Strong

200K context.

multilingual Moderate
speed Moderate
cost efficiency Moderate

Coding benchmarks

View the full coding benchmark leaderboard →
Benchmark Score Harness Source
swe-bench-verified ↗ 50.8% 2024-10 agentless 📊 leaderboard ↗
aider-polyglot ↗ 51.6% 2024-10 aider 📊 leaderboard ↗
livecodebench ↗ 38.9% 2025-01 — 📄 paper ↗
bigcodebench ↗ 25.7% 2025-04 — 📊 leaderboard ↗

Operator guidance

Retired: Anthropic retired Claude 3.5 Sonnet on 2025-10-28 and recommends Claude Sonnet 5.5 ($2 / $10 per 1M tokens) as the replacement (Anthropic model deprecations page, checked 2026-10-11).

Use cases

Limitations

Citations