SI Index Data as of Oct 9, 2026

Claude Haiku 4.5: Price, Context, Benchmarks, and Release Details

Anthropic provisional listing · first seen Oct 8, 2026 · score computed Oct 9, 2026, 01:50 UTC
60.5
SI Score (method si-v3-retained-evidence-2)
#62 of 140 ranked
64% confidence 64 percent, Medium confidence — 2 of 7 expected sources in
Coverage 51% of expected source weight · 100% confidence at 80% coverage

Pillar breakdown

Coding (weight 40 percent) —
Math (weight 15 percent) 78.2
Preference (weight 15 percent) 72.7
Reasoning (weight 30 percent) 65.8

Pillar weights: reasoning 30% · math 15% · coding 40% · preference 15%. Benchmark results use fixed 0–100 scales before averaging; incomplete evidence is shrunk toward 50.

Facts

Input price / 1M
$1.00models.devFirst-party hosted API; MIT models.dev transcription. Provider documentation: https://docs.anthropic.com/en/docs/about-claude/models. Exact canonical endpoint; lowest short-context Standard USD token tier; cache/batch discounts excluded. Deprecated endpoints excluded; anthropic/claude-haiku-4-5-20251001 Retrieved Oct 9, 2026 · MIT
Open source ↗
Output price / 1M
$5.00models.devFirst-party hosted API; MIT models.dev transcription. Provider documentation: https://docs.anthropic.com/en/docs/about-claude/models. Exact canonical endpoint; lowest short-context Standard USD token tier; cache/batch discounts excluded. Deprecated endpoints excluded; anthropic/claude-haiku-4-5-20251001 Retrieved Oct 9, 2026 · MIT
Open source ↗
Context window
200Kmodels.devPublished source fact Retrieved Oct 9, 2026 · MIT
Open source ↗
Max output
64Kmodels.devPublished source fact Retrieved Oct 9, 2026 · MIT
Open source ↗
Released
Oct 15, 2025models.devPublished source fact Retrieved Oct 9, 2026 · MIT
Open source ↗
Open weights
Nomodels.devPublished source fact Retrieved Oct 9, 2026 · MIT
Open source ↗
License
not yet reported
Input modalities
text, image, pdfmodels.devPublished source fact Retrieved Oct 9, 2026 · MIT
Open source ↗

Benchmark results

7 results
Benchmark Raw result Normalized (0–100) Pillar
gpqa diamond 60.5%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Oct 16, 2025 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		60.5 reasoning
gpqa diamond 71.2%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] 32KPublished Oct 22, 2025 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		71.2 reasoning
lmarena text 1396.1 eloLMArena / ArenaPublished source fact [variant] text / overallPublished Oct 2, 2026 Retrieved Oct 9, 2026 · CC-BY-4.0
Open source ↗ 		72.7 preference
math level 5 86.9%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Oct 16, 2025 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		86.9 math
math level 5 96.4%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] 32KPublished Oct 22, 2025 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		96.4 math
otis mock aime 2024 2025 35.8%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Oct 16, 2025 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		35.8 math
otis mock aime 2024 2025 66.7%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] 32KPublished Oct 22, 2025 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		66.7 math

Normalization uses fixed absolute 0–100 scales for each unit, independently of other models that have a result on each benchmark. Coverage and evidence breadth still affect the composite; compare the evaluation conditions before reading a small score gap as decisive. “lab-reported” marks the provider's own published figure.

Sources: in and pending

in Reported (2)

  • Epoch AI Benchmarking · arrived Oct 8, 2026
  • LMArena / Arena · arrived Oct 8, 2026

pending Awaiting (5)

  • ARC Prize · carries 13% of expected weight
  • Humanity’s Last Exam · carries 13% of expected weight
  • Official model cards via models.dev · carries 4% of expected weight
  • LiveBench · carries 13% of expected weight
  • Terminal-Bench · carries 6% of expected weight

The confidence % rises as pending sources publish. Some sources never cover some models — that is why 100% confidence arrives at 80% of expected weight, not at full coverage.