in the loop

Gemma 3 27B
#158 of 274 scored.

GoogleMar 12, 2025Open weights

Capabilities Index

Range 124.3–132.2

130.0

Context

Takes text, image

131K

Input

Per 1M tokens

$0.08

Output

Per 1M tokens

$0.45

Benchmarks

How it scores.

  • GPQA Diamond30.3%

    141st of 186

  • Mock AIME22.4%

    127th of 176

Everything else Epoch has for it

  • Lech Mazur Writing79.9%
  • MATH level 574.0%
  • GeoBench52.0%
  • Fiction.LiveBench33.3%
  • DTBench20.9%
  • LMCA14.4%
  • Aider polyglot4.9%
  • Chess Puzzles0.0%

On the index

Around it.

  1. #155Magistral Small 1.2Mistral131.4
  2. #156Grok-2 (Dec 2024)xAI130.5
  3. #157Phi-4Microsoft · $0.07 in130.4
  4. #158Gemma 3 27BGoogle · $0.08 in130.0
  5. #159Claude 3.5 SonnetAnthropic130.0
  6. #160Llama 4 ScoutMeta · $0.10 in129.6
  7. #161GPT-4.1 nanoOpenAI · $0.10 in129.6

Sources