Noometry

Model comparison

Claude Haiku 4.5 vs Llama 4 Maverick

Claude Haiku 4.5 is the stronger model overall, scoring 39.5 to 30.9 on the Noometry Index. Llama 4 Maverick costs 6.6× less per token, which makes it the better buy when Claude Haiku 4.5's lead doesn't matter for your workload.

Last verified . 40 shared benchmarks.

Claude Haiku 4.5 Anthropic

39.5

Rank #165 Confirmed

Llama 4 Maverick Meta

30.9

Rank #282 Confirmed

Summary

  • They share 40 benchmarks with published results for both. Claude Haiku 4.5 scores higher in 8 categories and Llama 4 Maverick in 2 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in writing & preference, where Claude Haiku 4.5 leads 57.9 to 38.8.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 66.7% for Claude Haiku 4.5 and 20.6% for Llama 4 Maverick.
  • Llama 4 Maverick is cheaper at $0.19 / $0.65 per million input/output tokens, against $1 / $5 for Claude Haiku 4.5.
  • Claude Haiku 4.5 accepts more context: 200K tokens versus 128K.
  • Llama 4 Maverick has downloadable open weights; the other is API-only.

Side by side

Claude Haiku 4.5Llama 4 Maverick
ProviderAnthropicMeta
Noometry Index39.530.9
Released2025-10-152025-04-05
WeightsProprietaryOpen
Context window200K128K
Max output64K4K
Input $ / M tokens$1$0.19
Output $ / M tokens$5$0.65
Results tracked5354

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

Category by category

Coding Claude Haiku 4.5 leads

Claude Haiku 4.5: 44.0 (#78), Llama 4 Maverick: 26.6 (#324)

BenchmarkClaude Haiku 4.5Llama 4 Maverick
SWE-bench Verified (bash only)66.6%21%
SciCode43.3%33.1%
WeirdML45.4%24.5%
LMArena Coding14531302
ALE-Bench653.48172.97
Aider Polyglot—15.6%
LMArena WebDev1330—
SWE-bench Multilingual64.7%—
BigCodeBench Instruct—49.7%
BigCodeBench Complete—61.4%

Agentic & Tool Use Claude Haiku 4.5 leads

Claude Haiku 4.5: 33.6 (#52), Llama 4 Maverick: 28.2 (#91)

BenchmarkClaude Haiku 4.5Llama 4 Maverick
Berkeley Function Calling Leaderboard68.7%37.3%
Terminal-Bench35.5%—
DeepResearch Bench45.5%—
BALROG31.2%—
ExploitBench13.7%—
Vending-Bench 2458.89—

Reasoning Claude Haiku 4.5 leads

Claude Haiku 4.5: 15.1 (#320), Llama 4 Maverick: 10.1 (#342)

BenchmarkClaude Haiku 4.5Llama 4 Maverick
ARC-AGI-24%0%
NYT Connections (extended)14.3%8%
ARC-AGI-147.7%4.4%
CritPt0%0%
LMArena Hard Prompts14201281
DTBench73.6%61.9%
LMCA30.9%15.9%
Epoch Capabilities Index142.41132.2
ForecastBench61.457.5
SimpleBench—27.7%
Kagi LLM Benchmark—55.9%
Chess Puzzles8%—
EnigmaEval—0.6%

Math Claude Haiku 4.5 leads

Claude Haiku 4.5: 44.9 (#78), Llama 4 Maverick: 26.0 (#262)

BenchmarkClaude Haiku 4.5Llama 4 Maverick
OTIS Mock AIME 2024-202566.7%20.6%
Omni-MATH56.1%42.2%
LMArena Math13961299
MATH Level 596.4%73%
FrontierMath (Feb 2025 set)5.9%0.7%
FrontierMath Tier 4 (v1)2.1%—

Knowledge Claude Haiku 4.5 leads

Claude Haiku 4.5: 37.7 (#153), Llama 4 Maverick: 33.4 (#204)

BenchmarkClaude Haiku 4.5Llama 4 Maverick
GPQA Diamond71.2%67%
MMLU-Pro77.7%81%
Vectara Hallucination Rate9.8%8.2%
GPQA (HELM)60.5%65%
LMArena Expert14421259
Humanity's Last Exam—5.7%
SimpleQA Verified13.2%—
Confabulations—22.6%

Multimodal Llama 4 Maverick leads

Claude Haiku 4.5: 26.8 (#118), Llama 4 Maverick: 31.6 (#105)

BenchmarkClaude Haiku 4.5Llama 4 Maverick
LMArena Vision—1142
GeoBench—52%
Blueprint-Bench 20%—
LMArena Document1420—
SpatialViz-Bench—31.8%

Multilingual Claude Haiku 4.5 leads

Claude Haiku 4.5: 49.9 (#129), Llama 4 Maverick: 42.2 (#195)

BenchmarkClaude Haiku 4.5Llama 4 Maverick
LMArena Non-English13771269
LMArena Chinese14171277
LMArena French14081259
LMArena German13751291
LMArena Japanese13391207
LMArena Korean13471203
LMArena Russian13811286
LMArena Spanish14201293

Instruction Following Too close to call

Claude Haiku 4.5: 71.4 (#149), Llama 4 Maverick: 71.7 (#146)

BenchmarkClaude Haiku 4.5Llama 4 Maverick
IFEval80.1%90.8%
LMArena Instruction Following14141267

Long Context Claude Haiku 4.5 leads

Claude Haiku 4.5: 43.6 (#92), Llama 4 Maverick: 31.4 (#279)

BenchmarkClaude Haiku 4.5Llama 4 Maverick
LMArena Longer Query14271280
Fiction.LiveBench—46.2%

Writing & Preference Claude Haiku 4.5 leads

Claude Haiku 4.5: 57.9 (#123), Llama 4 Maverick: 38.8 (#252)

BenchmarkClaude Haiku 4.5Llama 4 Maverick
LMArena Text13961287
LMArena Creative Writing13721267
WildBench83.9%80%
LMArena Multi-Turn14091289
Short-Story Creative Writing—62%
EQ-Bench Creative Writing—860
EQ-Bench 41064—

Frequently asked questions

Is Claude Haiku 4.5 better than Llama 4 Maverick?

Claude Haiku 4.5 is the stronger model overall, scoring 39.5 to 30.9 on the Noometry Index. Llama 4 Maverick costs 6.6× less per token, which makes it the better buy when Claude Haiku 4.5's lead doesn't matter for your workload.

Which is cheaper, Claude Haiku 4.5 or Llama 4 Maverick?

Llama 4 Maverick is cheaper. It lists at $0.19 per million input tokens and $0.65 per million output tokens; Claude Haiku 4.5 lists at $1 and $5.

Is Claude Haiku 4.5 or Llama 4 Maverick better for coding?

Claude Haiku 4.5 scores higher on coding benchmarks: 44.0 versus 26.6 in the Noometry coding category.

Which has the bigger context window?

Claude Haiku 4.5 does, with 200K tokens against 128K.

How many benchmarks do Claude Haiku 4.5 and Llama 4 Maverick share?

40 benchmarks have published results for both models. Claude Haiku 4.5 has 53 scored results on Noometry and Llama 4 Maverick has 54.

Related comparisons

Go deeper