Compare

Claude Haiku 4.5 vs Llama 4 Behemoth

Anthropic vs Meta. Specs, benchmarks, and real per-task cost — all in one page.

Verdict

Llama 4 Behemoth leads on LMSYS ELO (1342 vs 1305). Claude Haiku 4.5 is ~2.4× cheaper on a 3:1 input:output blend. Llama 4 Behemoth is open-weights, the other is proprietary.

Anthropic

Claude Haiku 4.5

Fast, cheap, and smart enough for most routing and extraction tasks. Great base model for subagents.

Sub-second latency10x cheaper than OpusStrong structured output

Meta

Llama 4 Behemoth

Meta’s open-weights flagship. 405B params, fully open license, runs on every major inference provider.

Best open-weights modelRuns anywhereTransparent training

Pricing

Input / 1M$0.80$2.50

Output / 1M$4.00$8.00

Context200K256K

Max output16K16K

LicenseProprietaryOpen weights

Released2025-10-012025-12-10

Benchmarks

LMSYS ELO1305.01342.0

MMLU Pro81.487.1

HumanEval87.089.3

SWE-Bench49.252.4

MATH76.881.5

GPQA—56.2

IFEval87.388.0

Per-task cost

Summarize a 1-hour meeting transcript$0.014$0.041

Review a 500-line pull request$0.014$0.036

Answer a customer support ticket$0.0056$0.014

Extract structured data from a resume$0.0080$0.021

Debug a stack trace with context$0.013$0.031

Per-call cost using published token counts for each task. Real-world prompts vary.

Tool

Interactive compare →

Add up to four models, tweak tasks live.