BlogLLM Benchmark Scores as Published, Side by Side

Gemini 2.5 Pro: Benchmark Scores as Published

Every result the catalog holds for Gemini 2.5 Pro: 6 tests in 4 areas, from 6 pages. Each was printed by the company named beside it. Each number is as printed; nothing is ranked or rescored.

Developer GoogleReleased 2025-03-25Context window 1.05MMax output 65.5KInput Text, image, video, audio, PDFOutput TextWeights ClosedData version 2026103001, as of 2026-10-30

Knowledge Work

TestValuePrinted bySetup as printed
Vals Finance Agent (version not stated)29.4%Anthropic, Claude Sonnet 4.5checked 2026-10-15-

Agentic Coding

TestValuePrinted bySetup as printed
SWE-bench Verified63.2%Anthropic, Claude Opus 4 and Claude Sonnet 4checked 2026-10-15; printed on 3 more pages-
Terminal-Bench (version not stated)25.3%Anthropic, Claude Opus 4 and Claude Sonnet 4checked 2026-10-15; printed on 3 more pages-

Science and Math

TestValuePrinted bySetup as printed
GPQA Diamond83%Anthropic, Claude Opus 4 and Claude Sonnet 4checked 2026-10-15; printed on 3 more pages-
AIME 202583%Anthropic, Claude Opus 4 and Claude Sonnet 4checked 2026-10-15; printed on 3 more pages-

Multimodal Understanding

TestValuePrinted bySetup as printed
MMMU (validation)79.6%Anthropic, Claude Opus 4 and Claude Sonnet 4checked 2026-10-15; printed on 3 more pages-

Price

API list prices in US dollars per one million tokens, as printed on the developer's pricing page on the day checked. Cost per task is measured by Vals AI.

PriceValuePrinted bySetup as printed
Input$1.25Google, Gemini API pricingchecked 2026-10-15Context: up to 200,000 tokens; Long context: $2.5 per 1M tokens above 200,000 tokens
Cached input$0.125Google, Gemini API pricingchecked 2026-10-15Context: up to 200,000 tokens; Long context: $0.25 per 1M tokens above 200,000 tokens
Output$10Google, Gemini API pricingchecked 2026-10-15Context: up to 200,000 tokens; Long context: $15 per 1M tokens above 200,000 tokens
Input, batch$0.625Google, Gemini API pricingchecked 2026-10-15Context: up to 200,000 tokens; Long context: $1.25 per 1M tokens above 200,000 tokens
Cached input, batch$0.125Google, Gemini API pricingchecked 2026-10-15Context: up to 200,000 tokens; Long context: $0.25 per 1M tokens above 200,000 tokens
Output, batch$5Google, Gemini API pricingchecked 2026-10-15Context: up to 200,000 tokens; Long context: $7.5 per 1M tokens above 200,000 tokens
Input, flex$0.625Google, Gemini API pricingchecked 2026-10-15Context: up to 200,000 tokens; Long context: $1.25 per 1M tokens above 200,000 tokens
Cached input, flex$0.125Google, Gemini API pricingchecked 2026-10-15Context: up to 200,000 tokens; Long context: $0.25 per 1M tokens above 200,000 tokens
Output, flex$5Google, Gemini API pricingchecked 2026-10-15Context: up to 200,000 tokens; Long context: $7.5 per 1M tokens above 200,000 tokens
Input, priority$2.25Google, Gemini API pricingchecked 2026-10-15Context: up to 200,000 tokens; Long context: $4.5 per 1M tokens above 200,000 tokens
Cached input, priority$0.225Google, Gemini API pricingchecked 2026-10-15Context: up to 200,000 tokens; Long context: $0.45 per 1M tokens above 200,000 tokens
Output, priority$18Google, Gemini API pricingchecked 2026-10-15Context: up to 200,000 tokens; Long context: $27 per 1M tokens above 200,000 tokens
Cost per taskVals AI, USD, lower is better$1.22Vals AI, Vibe Code Bench v1.1 leaderboardchecked 2026-10-27Harness: OpenHands

More from Google

Every number here is read from the same catalog file as the benchmark table, data version 2026103001. Where one page printed the model under several setups, each value is listed. The catalog is public in armanobosyan/sugra-llm-benchmarks; a wrong number can be reported from the table, with Report an error under the value, or as an issue in that repository.