Skip to content
Model Pareto

Claude Haiku 4.5

AnthropicProprietaryanthropic.com ↗
Compare with nearest rivalsCompare with…
Input
$1.00/M tok
Output
$5.00/M tok
Blended
$2.00/M tok · 3:1
Speed
109tok/s
Context
200Ktokens

Previous-generation small model.

Rankings

Benchmark scores

~ italic, dashed = no published score yet, estimated from related benchmarks.

Coding

SWE-bench Pro (Public)Estimated
~#18 of 19 (estimated)~51.0%
  1. #1Claude Opus 5.589.9%
  2. #2Claude Sonnet 5.581.3%
  3. #17Muse Glimmer51.2%
  4. ~#18Claude Haiku 4.5Estimated51.0%
  5. #18Nex-N2.5-Mini43.8%
No published score — estimated from related benchmarks
DeepSWE v1.1Estimated
~#28 of 31 (estimated)~48.3%
  1. #1DeepSeek V4.1 Flash74.2%
  2. #2Grok 4.772.6%
  3. #27Qwen3.8 Max (0902)51.0%
  4. ~#28Claude Haiku 4.5Estimated48.3%
  5. #28Qwen3.8 27B42.2%
No published score — estimated from related benchmarks
AA Coding Agent IndexEstimated
~last of 15 (estimated)below measured range (<41.1)
  1. #1Claude Opus 5.566.0
  2. #2Claude Fable 5.162.2
  3. #14GPT-6 Luna41.1
  4. ~#15Claude Haiku 4.5Estimated<41.1
No published score — estimated from related benchmarks

Writing & chat

LMArena Text, Non-English (Elo)Verified
#38 of 531395 Elo
  1. #1Claude Opus 5.51501
  2. #2Claude Fable 5.1Few votes1496
  3. #36Qwen3.5 122B A10B1397
  4. #38Claude Haiku 4.51395
  5. #39GPT-5.4 nano1387
lmarena.ai · 2026-09-24
LMArena Text (Elo)Verified
#37 of 531415 Elo
  1. #1Claude Opus 5.51509
  2. #2Claude Fable 5.11501
  3. #36Qwen3.5 122B A10B1417
  4. #37Claude Haiku 4.51415
  5. #38Mistral Large 31413
lmarena.ai · 2026-09-24
EQ-Bench Creative Writing v3 (Elo)Estimated
~#19 of 26 (estimated)~1585 Elo
  1. #1GPT-6 Astra2173
  2. #2Claude Fable 5.12162
  3. #18Inkling1611
  4. ~#19Claude Haiku 4.5Estimated1585
  5. #19Gemini 3.5 Flash-Lite1559
No published score — estimated from related benchmarks

Vision

CharXiv ReasoningEstimated
~last of 9 (estimated)below measured range (<74.5%)
  1. #1Gemini 3.8 Flash86.2%
  2. #2Kimi K384.8%
  3. #8Gemini 3.5 Flash-Lite74.5%
  4. ~#9Claude Haiku 4.5Estimated<74.5%
No published score — estimated from related benchmarks
ChartographyEstimated
~last of 23 (estimated)below measured range (<9.0%)
  1. #1GPT-6 Astra71.0%
  2. #2Claude Opus 5.566.3%
  3. #22Mistral Large 39.0%
  4. ~#23Claude Haiku 4.5Estimated<9.0%
No published score — estimated from related benchmarks

Hard reasoning

Agents

OSWorld 2.0Estimated
~last of 5 (estimated)below measured range (<47.5%)
  1. #1GLM-5.3 Flash59.1%
  2. #2Kimi K358.3%
  3. #4GPT-5.547.5%
  4. ~#5Claude Haiku 4.5Estimated<47.5%
No published score — estimated from related benchmarks
BrowseCompEstimated
~#15 of 18 (estimated)~56.7%
  1. #1Atria Dawn Preview92.5%
  2. #2GPT-6 Astra91.5%
  3. #14Inkling77.1%
  4. ~#15Claude Haiku 4.5Estimated56.7%
  5. #15Mistral Medium 3.548.6%
No published score — estimated from related benchmarks