AI Index

Claude Haiku 3.5

Language Anthropic Partly verified Retired Shut down Feb 19, 2026 Released Nov 4, 2024
Model ID
Compare with… →
Context
200K
8K max output
Input price
—
Not listed
Output price
—
Not listed
Epoch Capabilities Index
—
Not listed
LMArena Text
1,325
#135 of 149 listed

Where to run it

1 offering · prices as listed
MeterTierPrice
Claude API Official Retired Global · USD claude-3-5-haiku-20241022
InputStandard past price, from the vendor’s page archived Feb 19, 2026$0.80 / 1M tokens
Batch past price, from the vendor’s page archived Feb 19, 2026$0.40 / 1M tokens
OutputStandard past price, from the vendor’s page archived Feb 19, 2026$4.00 / 1M tokens
Batch past price, from the vendor’s page archived Feb 19, 2026$2.00 / 1M tokens
Cache readStandard past price, from the vendor’s page archived Feb 19, 2026$0.080 / 1M tokens
Cache write5-minute cache past price, from the vendor’s page archived Feb 19, 2026$1.00 / 1M tokens
1-hour cache past price, from the vendor’s page archived Feb 19, 2026$1.60 / 1M tokens

Changes

All changes →

No changes recorded yet.

Cited scores

From third-party leaderboards, not measured by us
  • LMArena Text 1,325

    #135 of 149 listed models · CI 1,321–1,328 · 68,873 votes

  • Epoch AI · GPQA Diamond 38.1%

    #117 of 128 listed models · ±2.64

  • LMArena Text · Chinese 1,322

    #138 of 149 listed models · CI 1,313–1,332 · 4,642 votes

  • LMArena Text · Coding 1,386

    #130 of 149 listed models · CI 1,380–1,392 · 11,120 votes

  • LMArena Vision 1,092

    #79 of 87 listed models · CI 1,076–1,108 · 1,400 votes

  • LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Oct 2, 2026
  • Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Oct 3, 2026
  • changes we made