Noometry

Model comparison

Claude Haiku 4.5 vs Claude Sonnet 5

Claude Sonnet 5 is the stronger model overall, scoring 54.6 to 39.5 on the Noometry Index. Claude Haiku 4.5 costs 2.0× less per token, which makes it the better buy when Claude Sonnet 5's lead doesn't matter for your workload.

Last verified . 35 shared benchmarks.

Claude Haiku 4.5 Anthropic

39.5

Rank #165 Confirmed

Claude Sonnet 5 Anthropic

54.6

Rank #29 Confirmed

Summary

  • They share 35 benchmarks with published results for both. Claude Haiku 4.5 scores higher in 0 categories and Claude Sonnet 5 in 10 categories; 10 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Claude Sonnet 5 leads 49.1 to 15.1.
  • The biggest single-benchmark swing is NYT Connections (extended): 14.3% for Claude Haiku 4.5 and 75.1% for Claude Sonnet 5.
  • Claude Haiku 4.5 is cheaper at $1 / $5 per million input/output tokens, against $2 / $10 for Claude Sonnet 5.
  • Claude Sonnet 5 accepts more context: 1M tokens versus 200K.

Side by side

Claude Haiku 4.5Claude Sonnet 5
ProviderAnthropicAnthropic
Noometry Index39.554.6
Released2025-10-152026-06-29
WeightsProprietaryProprietary
Context window200K1M
Max output64K128K
Input $ / M tokens$1$2
Output $ / M tokens$5$10
Results tracked5351

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding Claude Sonnet 5 leads

Claude Haiku 4.5: 44.0 (#78), Claude Sonnet 5: 55.5 (#26)

BenchmarkClaude Haiku 4.5Claude Sonnet 5
LMArena WebDev13301541
SciCode43.3%54.3%
WeirdML45.4%68.8%
LMArena Coding14531483
ALE-Bench653.481,463
DeepSWE—53.8%
FrontierCode—42.7%
SWE-bench Verified (bash only)66.6%—
CursorBench—34.1%
SWE-bench Multilingual64.7%—
GSO—37.3%

Agentic & Tool Use Claude Sonnet 5 leads

Claude Haiku 4.5: 33.6 (#52), Claude Sonnet 5: 42.8 (#18)

BenchmarkClaude Haiku 4.5Claude Sonnet 5
Vending-Bench 2458.896,378
Terminal-Bench35.5%—
APEX-Agents—54.5%
Berkeley Function Calling Leaderboard68.7%—
DeepResearch Bench45.5%—
BALROG31.2%—
ExploitBench13.7%—
GBAEval—65.3%
LMArena Search—1194

Reasoning Claude Sonnet 5 leads

Claude Haiku 4.5: 15.1 (#320), Claude Sonnet 5: 49.1 (#39)

BenchmarkClaude Haiku 4.5Claude Sonnet 5
NYT Connections (extended)14.3%75.1%
CritPt0%16.9%
Chess Puzzles8%35%
LMArena Hard Prompts14201461
DTBench73.6%92.5%
LMCA30.9%50%
Epoch Capabilities Index142.41156.21
ForecastBench61.461.1
ARC-AGI-24%—
SimpleBench—60.6%
ARC-AGI-147.7%—
Mystery Game Puzzles—35%
Surface Evolver Bench—60%
Bench to the Future 3—0.14

Math Claude Sonnet 5 leads

Claude Haiku 4.5: 44.9 (#78), Claude Sonnet 5: 66.2 (#27)

BenchmarkClaude Haiku 4.5Claude Sonnet 5
OTIS Mock AIME 2024-202566.7%94.7%
LMArena Math13961467
FrontierMath (Tiers 1-3)—65.6%
FrontierMath Tier 4—29.3%
ProofBench—77%
Omni-MATH56.1%—
MATH Level 596.4%—
FrontierMath (Feb 2025 set)5.9%—
FrontierMath Tier 4 (v1)2.1%—

Knowledge Claude Sonnet 5 leads

Claude Haiku 4.5: 37.7 (#153), Claude Sonnet 5: 55.6 (#47)

BenchmarkClaude Haiku 4.5Claude Sonnet 5
GPQA Diamond71.2%90.5%
SimpleQA Verified13.2%33.7%
LMArena Expert14421490
MMLU-Pro77.7%—
Vectara Hallucination Rate9.8%—
GPQA (HELM)60.5%—

Multimodal Claude Sonnet 5 leads

Claude Haiku 4.5: 26.8 (#118), Claude Sonnet 5: 42.4 (#31)

BenchmarkClaude Haiku 4.5Claude Sonnet 5
Blueprint-Bench 20%24.9%
LMArena Document14201466
LMArena Vision—1274

Multilingual Claude Sonnet 5 leads

Claude Haiku 4.5: 49.9 (#129), Claude Sonnet 5: 53.8 (#55)

BenchmarkClaude Haiku 4.5Claude Sonnet 5
LMArena Non-English13771431
LMArena Chinese14171477
LMArena French14081460
LMArena German13751440
LMArena Japanese13391422
LMArena Korean13471411
LMArena Russian13811451
LMArena Spanish14201437

Instruction Following Claude Sonnet 5 leads

Claude Haiku 4.5: 71.4 (#149), Claude Sonnet 5: 76.3 (#41)

BenchmarkClaude Haiku 4.5Claude Sonnet 5
LMArena Instruction Following14141452
IFEval80.1%—

Long Context Claude Sonnet 5 leads

Claude Haiku 4.5: 43.6 (#92), Claude Sonnet 5: 44.8 (#55)

BenchmarkClaude Haiku 4.5Claude Sonnet 5
LMArena Longer Query14271463

Writing & Preference Claude Sonnet 5 leads

Claude Haiku 4.5: 57.9 (#123), Claude Sonnet 5: 69.2 (#25)

BenchmarkClaude Haiku 4.5Claude Sonnet 5
LMArena Text13961442
LMArena Creative Writing13721416
EQ-Bench 410641236
LMArena Multi-Turn14091454
EQ-Bench Creative Writing—1794
WildBench83.9%—

Frequently asked questions

Is Claude Haiku 4.5 better than Claude Sonnet 5?

Claude Sonnet 5 is the stronger model overall, scoring 54.6 to 39.5 on the Noometry Index. Claude Haiku 4.5 costs 2.0× less per token, which makes it the better buy when Claude Sonnet 5's lead doesn't matter for your workload.

Which is cheaper, Claude Haiku 4.5 or Claude Sonnet 5?

Claude Haiku 4.5 is cheaper. It lists at $1 per million input tokens and $5 per million output tokens; Claude Sonnet 5 lists at $2 and $10.

Is Claude Haiku 4.5 or Claude Sonnet 5 better for coding?

Claude Sonnet 5 scores higher on coding benchmarks: 55.5 versus 44.0 in the Noometry coding category.

Which has the bigger context window?

Claude Sonnet 5 does, with 1M tokens against 200K.

How many benchmarks do Claude Haiku 4.5 and Claude Sonnet 5 share?

35 benchmarks have published results for both models. Claude Haiku 4.5 has 53 scored results on Noometry and Claude Sonnet 5 has 51.

Related comparisons

Go deeper