Skip to main content
Launching the open-source gateway

Claude Haiku 5.5 vs Claude Opus 4.8

SpecClaude Haiku 5.5Claude Opus 4.8
Respan Index57.2 (#16)55.6 (#19)
Blended price per 1M$0.1$10
Context window1M1M
Input / output per 1M$0.05 / $0.25$5 / $25
Max output128K128K
Inputtext, visiontext, vision
Released2026-10-072026-05-28

Respan Index by category

Each category is scored 0 to 100, and how far ahead the winner is. Leads under a point are close to even.

CategoryClaude Haiku 5.5Claude Opus 4.8
Coding55.1 Claude Haiku 5.539.2 Claude Opus 4.8
Agents66.0 Claude Haiku 5.564.6 Claude Opus 4.8
Reasoning65.0 Claude Haiku 5.564.8 Claude Opus 4.8
Knowledge & language23.6 Claude Haiku 5.552.4 Claude Opus 4.8
Human preference55.5 Claude Haiku 5.555.0 Claude Opus 4.8

How the Respan Index works

Benchmarks both report

Claude Opus 4.8 leads on 16 of the 26 benchmarks both report. Higher is better. Each score links to where it was published.

BenchmarkClaude Haiku 5.5Claude Opus 4.8
MirrorCode (Epoch AI)78.9%65.6%
APEX-Agents (Mercor)56.1%48.9%
Chess Puzzles (Epoch AI)55%52%
LMArena Agent (LMArena)0.01430.0776
LMArena Elo14251482
LMArena WebDev (LMArena)15891556
Terminal-Bench 4.0 (Vals AI)36.4%23.2%
Vending-Bench 2 (Andon Labs)3602.465787.43
EBR-bench (Epoch AI)43.8%28.6%
FrontierMath Tier 4 v2 (Epoch AI)46.3%56.1%
FrontierMath Tiers 1-3 v2 (Epoch AI)75.1%80%
Furniture Assembly (Epoch AI)47.5%42.5%
GPQA Diamond89.6%93.6%
LiveBench Agentic Coding (LiveBench)57.2%50.5%
LiveBench Coding (LiveBench)77.9%81.8%
LiveBench Data Analysis (LiveBench)72.8%66%
LiveBench Instruction Following (LiveBench)66.5%72%
LiveBench Language (LiveBench)63.5%79.7%
LiveBench Mathematics (LiveBench)92.8%94.3%
LiveBench Reasoning (LiveBench)84.6%89.2%
Mystery Game Puzzles (Epoch AI)30%36%
OTIS Mock AIME 2024-2025 (Epoch AI)98.9%98.3%
SAGE (Vals AI)45.8%54.8%
SimpleQA Verified23.8%53%
Humanity's Last Exam (no tools)45.9%49.8%
Humanity's Last Exam (with tools)57.4%57.9%

FAQ

Is Claude Haiku 5.5 better than Claude Opus 4.8?

Claude Haiku 5.5 scores 57.2 on the Respan Index (#16), ahead of Claude Opus 4.8 at 55.6 (#19). Claude Opus 4.8 leads on 16 of the 26 benchmarks both report.

Which is cheaper, Claude Haiku 5.5 or Claude Opus 4.8?

Claude Haiku 5.5 lists at $0.05 input and $0.25 output per 1M tokens; Claude Opus 4.8 at $5 and $25. Claude Haiku 5.5 is 100x cheaper per token at list price.

Which has the larger context window?

Both accept 1M tokens of context.

Compare other models

vs

Built for AI agents.
Break less.
Ship more.

Start for freeGet a demo