Claude Opus 4.6 vs Claude Sonnet 4.5
As of Oct 14, 2026, Claude Opus 4.6 leads Claude Sonnet 4.5 in 7 of 7 shared categories: General LLM, Reasoning, Math, Science, Coding › Web dev, Long context, and Factuality.
Claude Opus 4.6
Close. Near-SOTA in Reasoning, rank #2.
Claude Sonnet 4.5
It used to be. Was SOTA in General LLM.
| Category | Claude Opus 4.6 | Claude Sonnet 4.5 |
|---|---|---|
| General LLM | #4 · Near-SOTA · General LLM Margin 2.7210884353741562 LMArena leaderboard dataset · Oct 13, 2026 | #49 · Was SOTA · General LLM Margin 33.33333333333333 LMArena leaderboard dataset · Oct 13, 2026 |
| Reasoning | #2 · Near-SOTA · Reasoning Margin 2.0547945205479436 LMArena leaderboard dataset · Oct 13, 2026 | #40 · Was SOTA · Reasoning Margin 28.082191780821915 LMArena leaderboard dataset · Oct 13, 2026 |
| Math | #11 · #11 in Math Margin 22.22222222222223 Epoch AI benchmarking hub · Feb 5, 2026 | #25 · #25 in Math Margin 53.333333333333336 Epoch AI benchmarking hub · Sep 29, 2025 |
| Science | #32 · #32 in Science Margin — Epoch AI benchmarking hub · Feb 5, 2026 | #82 · #82 in Science Margin — Epoch AI benchmarking hub · Sep 29, 2025 |
| Coding | #— · Not enough data Margin — Epoch AI benchmarking hub · Feb 5, 2026 | #— · Not enough data Margin — Epoch AI benchmarking hub · Sep 29, 2025 |
| Coding › Web dev | #35 · Was SOTA · Coding › Web dev Margin 31.775700934579433 LMArena leaderboard dataset · Oct 13, 2026 | #78 · Was SOTA · Coding › Web dev Margin 71.96261682242991 LMArena leaderboard dataset · Oct 13, 2026 |
| Long context | #2 · Near-SOTA · Long context Margin 0.6802721088435391 LMArena leaderboard dataset · Oct 13, 2026 | #29 · Was SOTA · Long context Margin 19.72789115646259 LMArena leaderboard dataset · Oct 13, 2026 |
| Vision | #— · Not enough data Margin — LMArena leaderboard dataset · Oct 9, 2026 | — |
| Factuality | #3 · Near-SOTA · Factuality Margin 2.9850746268656536 LMArena leaderboard dataset · Oct 13, 2026 | #50 · #50 in Factuality Margin 38.059701492537314 LMArena leaderboard dataset · Oct 13, 2026 |
| Abstraction (ARC) | — | #66 · #66 in Abstraction (ARC) Margin — Epoch AI benchmarking hub · Sep 29, 2025 |
Claude Opus 4.6: $10 per 1M blended tokens.
Claude Sonnet 4.5: $6 per 1M blended tokens.
Claude Opus 4.6: closed weights.
Claude Sonnet 4.5: closed weights.
General LLM: As of Oct 14, 2026, Claude Opus 4.6 leads Claude Sonnet 4.5 in General LLM at #4 against #49.
Reasoning: As of Oct 14, 2026, Claude Opus 4.6 leads Claude Sonnet 4.5 in Reasoning at #2 against #40.
Math: As of Oct 14, 2026, Claude Opus 4.6 leads Claude Sonnet 4.5 in Math at #11 against #25.
Science: As of Oct 14, 2026, Claude Opus 4.6 leads Claude Sonnet 4.5 in Science at #32 against #82.
Coding › Web dev: As of Oct 14, 2026, Claude Opus 4.6 leads Claude Sonnet 4.5 in Coding › Web dev at #35 against #78.
Long context: As of Oct 14, 2026, Claude Opus 4.6 leads Claude Sonnet 4.5 in Long context at #2 against #29.
Factuality: As of Oct 14, 2026, Claude Opus 4.6 leads Claude Sonnet 4.5 in Factuality at #3 against #50.
Claude Opus 4.6 · Claude Sonnet 4.5
Claude Opus 4.6 or Claude Sonnet 4.5?
As of Oct 14, 2026, Claude Opus 4.6 leads Claude Sonnet 4.5 in 7 of 7 shared categories: General LLM, Reasoning, Math, Science, Coding › Web dev, Long context, and Factuality.
Claude Opus 4.6 or Claude Sonnet 4.5 for General LLM?
As of Oct 14, 2026, Claude Opus 4.6 leads Claude Sonnet 4.5 in General LLM at #4 against #49.
Claude Opus 4.6 or Claude Sonnet 4.5 for Reasoning?
As of Oct 14, 2026, Claude Opus 4.6 leads Claude Sonnet 4.5 in Reasoning at #2 against #40.
Claude Opus 4.6 or Claude Sonnet 4.5 for Math?
As of Oct 14, 2026, Claude Opus 4.6 leads Claude Sonnet 4.5 in Math at #11 against #25.