Compare

Gemini 2.5 Flash vs Claude Sonnet 4.6

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

Verdict

Claude Sonnet 4.6 leads on LMSYS ELO (1368 vs 1312). Gemini 2.5 Flash is ~11.4× cheaper on a 3:1 input:output blend.

Google DeepMind

Google’s price/performance darling. 1M context at $0.30/M input — nobody comes close on throughput-per-dollar.

Absurd cost-to-context ratioFast multimodalFree tier via AI Studio

Anthropic

The workhorse. Near-Opus quality at 1/5 the cost. The default choice for production code and agent workloads.

Excellent cost/quality ratioFast inferenceGreat for Cursor/Claude Code

Pricing

Input / 1M$0.30$3.00

Output / 1M$1.20$15.00

Context1.0M500K

Max output32K64K

LicenseProprietaryProprietary

Released2026-02-052025-11-04

Benchmarks

LMSYS ELO1312.01368.0

MMLU Pro82.188.7

HumanEval85.892.1

SWE-Bench—64.5

MATH84.285.0

GPQA—59.4

MMMU74.674.1

IFEval86.190.5

Per-task cost

Summarize a 1-hour meeting transcript$0.0051$0.052

Review a 500-line pull request$0.0048$0.054

Answer a customer support ticket$0.0019$0.021

Extract structured data from a resume$0.0027$0.030

Debug a stack trace with context$0.0042$0.048

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

Add up to four models, tweak tasks live.