Models

Google

Gemma 3 27B Instruct

GoogleGoogleUnited States

Gemma 3 27B Instruct is a 27-billion parameter open-source instruction-tuned model from Google, built on the Gemma 3 architecture. It is designed for strong reasoning and instruction-following capabilities, suitable for a wide range of text generation tasks. The model supports a long context window, making it effective for processing lengthy documents.

ReasoningLong contextCodingCheap
Input / 1M tokens
$0.11
Output / 1M tokens
$0.25
Output tokens/s
23.92
First-token seconds
0.76s
Supported plans
0

Benchmark history

Evaluations

15

TAU2

Measured May 14, 2026Source

Score

0.11

Terminalbench Hard

Measured May 14, 2026Source

Score

0.04

Lcr

Measured May 14, 2026Source

Score

0.06

Ifbench

Measured May 14, 2026Source

Score

0.32

Aime 25

Measured May 14, 2026Source

Score

0.21

Aime

Measured May 14, 2026Source

Score

0.25

Math 500

Measured May 14, 2026Source

Score

0.88

Scicode

Measured May 14, 2026Source

Score

0.21

Livecodebench

Measured May 14, 2026Source

Score

0.14

Hle

Measured May 14, 2026Source

Score

0.05

Gpqa

Measured May 14, 2026Source

Score

0.43

Mmlu Pro

Measured May 14, 2026Source

Score

0.67

Artificial Analysis Math Index

Measured May 14, 2026Source

Score

20.7

Artificial Analysis Coding Index

Measured May 14, 2026Source

Score

9.6

Artificial Analysis Intelligence Index

Measured May 14, 2026Source

Score

10.3

Plan availability

Products and plans that support this model

0
No products or plans have been linked to this model yet.

Discussion

Thinking... Make sure you are connected to GitHub server