Anthropic · Benchmarks

Claude Haiku 5.5 benchmarks

Claude Haiku 5.5 has 5 published scores in this dataset, from a Anthropic release dated 2026-10-07. Its composite index is 72.4, which places it 20th of 22 ranked models. Its strongest area is agentic (index 82.0) and its weakest is coding (index 63.8). Its best placing is on Humanity's Last Exam, where its 45.9 ranks 10th of 23 models with a published score.

Closed APIOfficial page

Category profile

Index 0–100: each score relative to the best published score on that test.

ReasoningCodingAgenticKnowledge
Reasoning
71.31 test · not ranked
Coding
63.8#13 of 13 · 2 tests
Agentic
82.01 test · not ranked
Knowledge
—no data

Model facts

Lab
Anthropic
Type
Closed API
Released
2026-10-07
Context
1M
API price / 1M tokens
$0.1 in · $0.5 out

Claude Haiku 5.5 pricing and open alternatives

Claude Haiku 5.5 is only available through Anthropic's API, so it cannot be self-hosted; it costs $0.1 per million input tokens and $0.5 per million output tokens.

Open-weight alternatives

The highest-ranked open models, which you can run on your own GPU.

Claude Haiku 5.5 speed and pricing

Measured figures link to their source; estimates are labelled.

API output speed
236 tok/sArtificial Analysis median · Artificial Analysis
Time to first token
1.51 sOpenRouter model page provider stats
API price per 1M tokens
$0.100 in · $0.500 outlab list price · Anthropic pricing docs
Compare speed and price across all models

Score by test

Rank is among every model with a published score on that test.

Humanity's Last Exam45.9%Humanity's Last Exam, no tools; max effort#10 of 23 · 18.5 behind Claude Opus 5.5Humanity's Last Exam (with tools)57.4%Humanity's Last Exam, with tools; max effort#9 of 17 · 12.6 behind Kimi K3.1 ProSWE-bench Pro64.8%SWE-bench Pro; adaptive thinking max effort; avg of 5 trials#6 of 9 · 25.1 behind Claude Opus 5.5Terminal-Bench 4.039.2%Terminal-Bench 4.0; safeguards enabled#10 of 16 · 31.4 behind Claude Sonnet 5.5LMArena Text Elo1425LMArena Text, Overall; snapshot Oct 27, 2026; Claude Haiku 5.5 (High); +/-7#21 of 21 · 95 behind Gemini 4 Argon

Nearby in the composite ranking

Models with the closest composite index.

  1. 1Step 5 Preview78.7
  2. 2Nemotron 3 Ultra74.4
  3. 3Gemma 4 26B A4B72.9
  4. 4Claude Haiku 5.572.4
  5. 5Gemini 3.8 Flash70.8
  6. 6gpt-oss-120b57.3

Head-to-head comparisons

Side-by-side pages built from the tests both models report.

No head-to-head page for Claude Haiku 5.5 yet. Compare it on the rankings dashboard.

Sources

ModelTestScoreSettingSource
Claude Haiku 5.5Humanity's Last Exam45.9Humanity's Last Exam, no tools; max effortAnthropic Claude Haiku 5.5 System Card (Table 8.1.A); also in launch post
Claude Haiku 5.5Humanity's Last Exam (with tools)57.4Humanity's Last Exam, with tools; max effortAnthropic Claude Haiku 5.5 System Card (Table 8.1.A); also in launch post
Claude Haiku 5.5SWE-bench Pro64.8SWE-bench Pro; adaptive thinking max effort; avg of 5 trialsAnthropic Claude Haiku 5.5 System Card (Table 8.1.A / sec 8.2)
Claude Haiku 5.5Terminal-Bench 4.039.2Terminal-Bench 4.0; safeguards enabledAnthropic Claude Haiku 5.5 System Card (sec 8.4); also in launch post
Claude Haiku 5.5LMArena Text Elo1425LMArena Text, Overall; snapshot Oct 27, 2026; Claude Haiku 5.5 (High); +/-7LMArena (arena.ai) Text leaderboard

Scores are published results from model cards, lab reports, and public leaderboards, each linked to its source. Settings such as thinking mode or test version differ between sources and are shown next to each score. OpenLLM Buddy catalog models are uncensored fine-tunes; their scores are those of the official base model. Data collected 2026-10-30. How the index works