Compare

Claude Haiku 4.5 vs Gemini 2.5 Ultra

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

Verdict

Gemini 2.5 Ultra leads on LMSYS ELO (1385 vs 1305). Claude Haiku 4.5 is ~6.6× cheaper on a 3:1 input:output blend.

Anthropic

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

Sub-second latency10x cheaper than OpusStrong structured output

Google DeepMind

2M-token context, native video understanding, and Google’s deepest multimodal stack. The long-context king.

2M context windowNative videoBest multimodal reasoning

Pricing

Input / 1M$0.80$7.00

Output / 1M$4.00$21.00

Context200K2.0M

Max output16K64K

LicenseProprietaryProprietary

Released2025-10-012026-02-05

Benchmarks

LMSYS ELO1305.01385.0

MMLU Pro81.490.4

HumanEval87.090.2

SWE-Bench49.258.3

MATH76.892.0

MMMU—82.1

IFEval87.389.7

Per-task cost

Summarize a 1-hour meeting transcript$0.014$0.115

Review a 500-line pull request$0.014$0.098

Answer a customer support ticket$0.0056$0.038

Extract structured data from a resume$0.0080$0.056

Debug a stack trace with context$0.013$0.084

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

Add up to four models, tweak tasks live.