Claude Haiku 4.5
- Input
- $1.00/M tok
- Output
- $5.00/M tok
- Blended
- $2.00/M tok · 3:1
- Speed
- 109tok/s
- Context
- 200Ktokens
- Cost per task
- ~$2.1/taskest.
Previous-generation small model.
Rankings
Coding
Mixed#94 / 137rank by quality35.3±5.3quality / 1004/7 benchmarks measuredWriting & chat
Mixed#78 / 162rank by quality47.6±3.8quality / 1004/5 benchmarks measuredVision
Mixed#73 / 98rank by quality35.6±4.2quality / 1004/6 benchmarks measuredHard reasoning
Verified#116 / 218rank by quality39.7±3.3quality / 1004/4 benchmarks measuredAgents
Mixed#122 / 187rank by quality22.2±3.5quality / 1005/7 benchmarks measured
Benchmark scores
~ italic, dashed = no published score yet, estimated from related benchmarks.
Coding
SWE-bench VerifiedLab-reported
#13 of 2073.3%
- #1Claude Opus 596.0%
- #2Claude Sonnet 585.2%
- #12Flint74.2%
- #13Claude Haiku 4.573.3%
- #14Nemotron 3.5 Super VL71.1%
anthropic.com · 2025-10-15
SWE-bench Pro (Public)Estimated
~#23 of 24 (estimated)~49.8%
- #1Claude Fable 5.293.0%
- #2Claude Opus 5.589.9%
- #22Muse Glimmer51.2%
- ~#23Claude Haiku 4.5Estimated49.8%
- #23Nex-N2.5-Mini43.8%
No published score — estimated from related benchmarks
DeepSWE v1.1Estimated
~#41 of 46 (estimated)~42.2%
- #1Gemini 4 Argon78.8%
- #2DeepSeek V4.1 Flash 101777.0%
- #40Qwen3.8 27B42.2%
- ~#41Claude Haiku 4.5Estimated42.2%
- #41Laguna S 2.140.4%
No published score — estimated from related benchmarks
Terminal-Bench 4.0Verified
#63 of 1010.0%
- #1Kimi K3.1 Pro67.2%
- #2Claude Sonnet 5.563.6%
- #63Celeris-10.0%
- #63Claude Haiku 4.50.0%
- #63Devstral 20.0%
artificialanalysis.ai · 2026-09-24
LiveCodeBenchVerified
#16 of 2761.5%
- #1Qwen3.8-Omni-Flash92.6%
- #2Qwen3.8 Flash-Next91.9%
- #15Mistral Small 464.0%
- #16Claude Haiku 4.561.5%
- #17GLM-5.3 NVFP453.5%
artificialanalysis.ai · 2026-09-24
SciCodeVerified
#71 of 11842.2%
- #1Claude Fable 5.267.7%
- #2Claude Opus 5.566.9%
- #70Ling-3.0-flash-Fin42.4%
- #71Claude Haiku 4.542.2%
- #71Motif 342.2%
artificialanalysis.ai · 2026-09-24
AA Coding Agent IndexEstimated
~#25 of 26 (estimated)~29.5
- #1Claude Opus 5.561.7
- #2Claude Opus 559.7
- #24GPT-6 Luna31.1
- ~#25Claude Haiku 4.5Estimated29.5
- #25Muse Spark Open26.3
No published score — estimated from related benchmarks
Writing & chat
IFBenchVerified
#69 of 13954.3%
- #1MiniMax-M382.9%
- #2Qwen3.8 2.4T A95B82.8%
- #68Falcon-H1R-7B54.4%
- #69Claude Haiku 4.554.3%
- #70Jamba Reasoning 3B52.4%
artificialanalysis.ai · 2026-09-24
LMArena Text, Non-English (Elo)Verified
#41 of 561395 Elo
- #1Gemini 4 ArgonFew votes1512
- #2Claude Opus 5.51497
- #39Qwen3.5 122B A10B1397
- #41Claude Haiku 4.51395
- #42GPT-5.4 nano1387
lmarena.ai · 2026-09-24
LMArena Text (Elo)Verified
#40 of 561415 Elo
- #1Gemini 4 Argon1525
- #2Claude Opus 5.51507
- #39Qwen3.5 122B A10B1417
- #40Claude Haiku 4.51415
- #41Mistral Large 31413
lmarena.ai · 2026-09-24
EQ-Bench Creative Writing v3 (Elo)Estimated
~#19 of 26 (estimated)~1581 Elo
- #1GPT-6 Astra2173
- #2Claude Fable 5.12162
- #18Inkling1611
- ~#19Claude Haiku 4.5Estimated1581
- #19Gemini 3.5 Flash-Lite1559
No published score — estimated from related benchmarks
Vision
MMMU-ProVerified
#64 of 8758.6%
- #1Claude Opus 5.587.7%
- #2GPT-6 Astra86.9%
- #63Bonsai 27B60.5%
- #64Claude Haiku 4.558.6%
- #65Mistral Small 456.8%
artificialanalysis.ai · 2026-09-24
CharXiv ReasoningEstimated
~last of 12 (estimated)below measured range (<74.5%)
- #1Gemini 3.8 Flash86.2%
- #2Kimi K384.8%
- #11Gemini 3.5 Flash-Lite74.5%
- ~#12Claude Haiku 4.5Estimated<74.5%
No published score — estimated from related benchmarks
GDP.pdf (AA)Verified
#38 of 473.8%
- #1Muse Wonder Full32.6%
- #2GPT-6 Astra31.0%
- #37Qwen3.6 35B A3B6.6%
- #38Claude Haiku 4.53.8%
- #39Qwen3.5 122B A10B3.4%
artificialanalysis.ai · 2026-09-24
ChartographyEstimated
~last of 25 (estimated)below measured range (<9.0%)
No published score — estimated from related benchmarks
Hard reasoning
GPQA DiamondVerified
#120 of 19267.2%
- #1GPT-6 Astra96.1%
- #2Gemini 3.8 Flash95.3%
- #119Mistral Large 368.0%
- #120Claude Haiku 4.567.2%
- #121Llama 4 Maverick67.1%
artificialanalysis.ai · 2026-09-24
Humanity's Last ExamVerified
#117 of 19810.4%
- #1Claude Fable 5.274.3%
- #2Claude Opus 5.561.4%
- #116K2-V210.5%
- #117Claude Haiku 4.510.4%
- #117Solar Open 100B10.4%
artificialanalysis.ai · 2026-09-24
Agents
τ²-bench (Telecom)Verified
#49 of 11954.7%
- #1Step 3.7 Flash98.5%
- #2DeepSeek-V4-Pro-0813-nvfp4-DSpark98.3%
- #48Gemma 4 31B59.9%
- #49Claude Haiku 4.554.7%
- #50Nemotron Cascade 2 30B A3B53.2%
artificialanalysis.ai · 2026-09-24
OSWorld 2.0Estimated
~last of 5 (estimated)below measured range (<47.5%)
No published score — estimated from related benchmarks
BrowseCompEstimated
~#19 of 22 (estimated)~60.0%
- #1Atria Dawn Preview92.5%
- #2GPT-6 Astra91.5%
- #18Nemotron 3.5 Bolt 74B-A9B70.0%
- ~#19Claude Haiku 4.5Estimated60.0%
- #19Mistral Medium 3.548.6%
No published score — estimated from related benchmarks
GDPval-AA (Elo)Verified
#59 of 91719 Elo
- #1Claude Opus 5.51866
- #2Muse Wonder Full1857
- #58KAT Coder Pro V2720
- #59Claude Haiku 4.5719
- #60MiniCPM5-2B694
artificialanalysis.ai · 2026-09-24