Noometry

Model comparison

Claude Haiku 4.5 vs GPT-5.5

GPT-5.5 is the stronger model overall, scoring 63.4 to 39.3 on the Noometry Index. Claude Haiku 4.5 costs 5.6× less per token, which makes it the better buy when GPT-5.5's lead doesn't matter for your workload.

Last verified . 43 shared benchmarks.

Claude Haiku 4.5 Anthropic

39.3

Rank #155 Confirmed

GPT-5.5 OpenAI

63.4

Rank #8 Confirmed

Summary

  • They share 43 benchmarks with published results for both. Claude Haiku 4.5 scores higher in 0 categories and GPT-5.5 in 10 categories; 10 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where GPT-5.5 leads 72.7 to 15.1.
  • The biggest single-benchmark swing is NYT Connections (extended): 14.3% for Claude Haiku 4.5 and 96.2% for GPT-5.5.
  • Claude Haiku 4.5 is cheaper at $1 / $5 per million input/output tokens, against $5 / $30 for GPT-5.5.
  • GPT-5.5 accepts more context: 1.05M tokens versus 200K.

Side by side

Claude Haiku 4.5GPT-5.5
ProviderAnthropicOpenAI
Noometry Index39.363.4
Released2025-10-152026-04-23
WeightsProprietaryProprietary
Context window200K1.05M
Max output64K128K
Input $ / M tokens$1$5
Output $ / M tokens$5$30
Results tracked5371

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

Category by category

Coding GPT-5.5 leads

Claude Haiku 4.5: 43.9 (#71), GPT-5.5: 58.3 (#13)

BenchmarkClaude Haiku 4.5GPT-5.5
LMArena WebDev13301512
SciCode43.3%56.1%
WeirdML45.4%84.9%
LMArena Coding14541494
ALE-Bench653.481,943
SWE-bench Verified—80.6%
DeepSWE—67%
FrontierCode—43%
SWE-bench Verified (bash only)66.6%—
SWE-bench Multilingual64.7%—
GSO—40.2%
MirrorCode—10%

Agentic & Tool Use GPT-5.5 leads

Claude Haiku 4.5: 33.5 (#50), GPT-5.5: 50.7 (#6)

BenchmarkClaude Haiku 4.5GPT-5.5
Terminal-Bench35.5%84.7%
DeepResearch Bench45.5%54%
ExploitBench13.7%47.4%
Vending-Bench 2458.897,524
APEX-Agents—55.1%
Berkeley Function Calling Leaderboard68.7%—
OSWorld 2.0—13%
Remote Labor Index—6.3%
τ²-bench Banking—44.6%
PostTrainBench—27.2%
BALROG31.2%—
GBAEval—53.2%
GDP.pdf—26%
LMArena Search—1242

Reasoning GPT-5.5 leads

Claude Haiku 4.5: 15.1 (#310), GPT-5.5: 72.7 (#8)

BenchmarkClaude Haiku 4.5GPT-5.5
ARC-AGI-24%85%
NYT Connections (extended)14.3%96.2%
ARC-AGI-147.7%95%
CritPt0%27.1%
Chess Puzzles8%54%
LMArena Hard Prompts14201489
DTBench73.6%96%
LMCA30.9%54.3%
Epoch Capabilities Index142.42159.25
ForecastBench61.460.6
SimpleBench—69%
Kagi LLM Benchmark—88.8%
EBR-Bench—34.3%
Mystery Game Puzzles—56%
Surface Evolver Bench—88.1%
Bench to the Future 3—0.14

Math GPT-5.5 leads

Claude Haiku 4.5: 43.0 (#69), GPT-5.5: 81.5 (#9)

BenchmarkClaude Haiku 4.5GPT-5.5
OTIS Mock AIME 2024-202566.7%100%
LMArena Math13951486
FrontierMath (Feb 2025 set)5.9%51.7%
FrontierMath Tier 4 (v1)2.1%35.4%
FrontierMath (Tiers 1-3)—85.3%
FrontierMath Tier 4—72.5%
MathArena Final-Answer Competitions—94.3%
ProofBench—50%
Omni-MATH56.1%—
MATH Level 596.4%—
FrontierMath Erdős—0%

Knowledge GPT-5.5 leads

Claude Haiku 4.5: 37.6 (#146), GPT-5.5: 64.4 (#14)

BenchmarkClaude Haiku 4.5GPT-5.5
GPQA Diamond71.2%94%
SimpleQA Verified13.2%63%
Vectara Hallucination Rate9.8%9.3%
LMArena Expert14431508
MMLU-Pro77.7%—
GPQA (HELM)60.5%—

Multimodal GPT-5.5 leads

Claude Haiku 4.5: 27.0 (#111), GPT-5.5: 47.1 (#11)

BenchmarkClaude Haiku 4.5GPT-5.5
Blueprint-Bench 20%36.2%
LMArena Document14201486
LMArena Vision—1297
Furniture Assembly—44.2%

Multilingual GPT-5.5 leads

Claude Haiku 4.5: 50.1 (#124), GPT-5.5: 56.5 (#19)

BenchmarkClaude Haiku 4.5GPT-5.5
LMArena Non-English13761467
LMArena Chinese14151535
LMArena French14081486
LMArena German13771478
LMArena Japanese13391498
LMArena Korean13461461
LMArena Russian13801473
LMArena Spanish14201469

Instruction Following GPT-5.5 leads

Claude Haiku 4.5: 71.6 (#144), GPT-5.5: 77.7 (#18)

BenchmarkClaude Haiku 4.5GPT-5.5
LMArena Instruction Following14141479
IFEval80.1%—

Long Context GPT-5.5 leads

Claude Haiku 4.5: 43.6 (#86), GPT-5.5: 48.3 (#12)

BenchmarkClaude Haiku 4.5GPT-5.5
LMArena Longer Query14271484
CL-bench Life—22.2%

Writing & Preference GPT-5.5 leads

Claude Haiku 4.5: 58.1 (#119), GPT-5.5: 72.8 (#13)

BenchmarkClaude Haiku 4.5GPT-5.5
LMArena Text13961471
LMArena Creative Writing13711455
EQ-Bench 410641315
LMArena Multi-Turn14091477
EQ-Bench Creative Writing—1844
WildBench83.9%—

Frequently asked questions

Is Claude Haiku 4.5 better than GPT-5.5?

GPT-5.5 is the stronger model overall, scoring 63.4 to 39.3 on the Noometry Index. Claude Haiku 4.5 costs 5.6× less per token, which makes it the better buy when GPT-5.5's lead doesn't matter for your workload.

Which is cheaper, Claude Haiku 4.5 or GPT-5.5?

Claude Haiku 4.5 is cheaper. It lists at $1 per million input tokens and $5 per million output tokens; GPT-5.5 lists at $5 and $30.

Is Claude Haiku 4.5 or GPT-5.5 better for coding?

GPT-5.5 scores higher on coding benchmarks: 58.3 versus 43.9 in the Noometry coding category.

Which has the bigger context window?

GPT-5.5 does, with 1.05M tokens against 200K.

How many benchmarks do Claude Haiku 4.5 and GPT-5.5 share?

43 benchmarks have published results for both models. Claude Haiku 4.5 has 53 scored results on Noometry and GPT-5.5 has 71.

Related comparisons

Go deeper