Claude Haiku 4.5 vs GPT-5.4
As of Oct 15, 2026, Claude Haiku 4.5 leads GPT-5.4 in 0 of 7 shared categories.
Claude Haiku 4.5
#78 Best recorded rank: #78 in Abstraction (ARC), as of Oct 15, 2026.
GPT-5.4
It used to be. Was SOTA in General LLM.
| Category | Claude Haiku 4.5 | GPT-5.4 |
|---|---|---|
| General LLM | #133 · #133 in General LLM Margin 70.74468085106383 LMArena leaderboard dataset · Oct 13, 2026 | #30 · Was SOTA · General LLM Margin 15.957446808510639 LMArena leaderboard dataset · Oct 13, 2026 |
| Reasoning | #119 · #119 in Reasoning Margin 63.829787234042556 LMArena leaderboard dataset · Oct 13, 2026 | #32 · #32 in Reasoning Margin 17.55319148936171 LMArena leaderboard dataset · Oct 13, 2026 |
| Math | #— · Not enough data Margin — LMArena leaderboard dataset · Oct 13, 2026 | #9 · #9 in Math Margin 17.777777777777757 Epoch AI benchmarking hub · Mar 5, 2026 |
| Science | #109 · #109 in Science Margin — Epoch AI benchmarking hub · Oct 15, 2025 | #16 · #16 in Science Margin — Epoch AI benchmarking hub · Mar 5, 2026 |
| Coding | #— · Not enough data Margin — Scale Labs leaderboards · Oct 14, 2026 | #— · Not enough data Margin — LMArena leaderboard dataset · Oct 13, 2026 |
| Coding › Web dev | #103 · #103 in Coding › Web dev Margin 81.6 LMArena leaderboard dataset · Oct 13, 2026 | #60 · #60 in Coding › Web dev Margin 47.2 LMArena leaderboard dataset · Oct 13, 2026 |
| Long context | #107 · #107 in Long context Margin 56.91489361702128 LMArena leaderboard dataset · Oct 13, 2026 | #40 · #40 in Long context Margin 21.276595744680847 LMArena leaderboard dataset · Oct 13, 2026 |
| Vision | — | #— · Not enough data Margin — LMArena leaderboard dataset · Oct 9, 2026 |
| Factuality | #125 · #125 in Factuality Margin 73.25581395348837 LMArena leaderboard dataset · Oct 13, 2026 | #6 · Was SOTA · Factuality Margin 4.0697674418604635 LMArena leaderboard dataset · Oct 13, 2026 |
| Abstraction (ARC) | #78 · #78 in Abstraction (ARC) Margin — Epoch AI benchmarking hub · Oct 15, 2025 | #10 · #10 in Abstraction (ARC) Margin — Epoch AI benchmarking hub · Mar 5, 2026 |
Claude Haiku 4.5: $2.0000000000000004 per 1M blended tokens.
GPT-5.4: $5.625 per 1M blended tokens.
Claude Haiku 4.5: closed weights.
GPT-5.4: closed weights.
General LLM: As of Oct 15, 2026, GPT-5.4 leads Claude Haiku 4.5 in General LLM at #30 against #133.
Reasoning: As of Oct 15, 2026, GPT-5.4 leads Claude Haiku 4.5 in Reasoning at #32 against #119.
Science: As of Oct 15, 2026, GPT-5.4 leads Claude Haiku 4.5 in Science at #16 against #109.
Coding › Web dev: As of Oct 15, 2026, GPT-5.4 leads Claude Haiku 4.5 in Coding › Web dev at #60 against #103.
Long context: As of Oct 15, 2026, GPT-5.4 leads Claude Haiku 4.5 in Long context at #40 against #107.
Factuality: As of Oct 15, 2026, GPT-5.4 leads Claude Haiku 4.5 in Factuality at #6 against #125.
Abstraction (ARC): As of Oct 15, 2026, GPT-5.4 leads Claude Haiku 4.5 in Abstraction (ARC) at #10 against #78.
Claude Haiku 4.5 or GPT-5.4?
As of Oct 15, 2026, Claude Haiku 4.5 leads GPT-5.4 in 0 of 7 shared categories.
Claude Haiku 4.5 or GPT-5.4 for General LLM?
As of Oct 15, 2026, GPT-5.4 leads Claude Haiku 4.5 in General LLM at #30 against #133.
Claude Haiku 4.5 or GPT-5.4 for Reasoning?
As of Oct 15, 2026, GPT-5.4 leads Claude Haiku 4.5 in Reasoning at #32 against #119.
Claude Haiku 4.5 or GPT-5.4 for Science?
As of Oct 15, 2026, GPT-5.4 leads Claude Haiku 4.5 in Science at #16 against #109.