Skip to main content

Gemma 3 27B Instruct

GoogleGemmaOpen WeightGemma · Commercial OK

Description

Gemma 3 27B is a 27-billion-parameter vision-language model from Google, handling text and image input and generating text output. It features a 128K context window, multilingual support, and open weights. Suitable for complex question answering, summarization, reasoning, and image understanding tasks.

Release Date
2025-03-12
Parameters
27.0B
Context Length
262K
Modalities
image, text

Capability Radar

25
general
12
coding
34
reasoning
28
science
27
agents
70
multimodal

Rankings

Domain#RankScoreSource
Code Ranking495
11.0
AA
General Ranking464
26.0
AA
Multimodal Ranking54
44.0
LS
Science461
27.0
AA

Benchmark Scores (LLM Stats)

(LLM Stats (zeroeval))

Biology

GPQANYU + Cohere + Anthropic (2023)42.4%SR

Code

HumanEvalOpenAI (2021)87.8%SR
LiveCodeBench29.7%SR

Factuality

FACTS Grounding74.9%SR
SimpleQA10.0%SR

Finance

MMLU-Pro67.5%SR

General

IFEvalGoogle Research (2023)90.4%SR
Natural2Code84.5%SR
Global-MMLU-Lite75.1%SR
MBPP0.74 / 100SR
MMMU (val)64.9%SR
BIG-Bench Extra Hard19.3%SR
MRCR v2 (8-needle)13.5%SR

Image To Text

DocVQADocVQA (2020)86.6%SR
VQAv2 (val)71.0%SR
TextVQA65.1%SR

Language

BIG-Bench Hard87.6%SR
WMT24++53.4%SR
ECLeKTic16.7%SR

Math

GSM8k95.9%SR
MATH89.0%SR
MathVista-Mini67.6%SR
HiddenMath60.3%SR

Multimodal

AI2D84.5%SR
ChartQAMasry et al. (2022)78.0%SR
InfoVQA70.6%SR

Reasoning

Bird-SQL (dev)54.4%SR

AA Evaluation Indices

(Artificial Analysis)
Math Index(Artificial Analysis)
20.7
Coding Index(Artificial Analysis)
10.1
Intelligence Index(Artificial Analysis)
7.4
Math 500(OpenAI (2024), subset of Hendrycks et al. MATH (2021))
0.9
Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))
0.7
Gpqa(NYU + Cohere + Anthropic (2023))
0.4
Ifbench(Google Research (2023))
0.3
Aime(MAA (Mathematical Association of America))
0.3
Scicode(UIUC + Argonne National Lab (2024))
0.2
Aime 25(MAA (Mathematical Association of America))
0.2
Livecodebench(UC Berkeley + MIT + Cornell (2024))
0.1
Tau2(Sierra + U Toronto + Vector Institute (2025))
0.1
Lcr(Artificial Analysis)
0.1
Terminalbench V2 1
0.0
Hle(Center for AI Safety + Scale AI (2025))
0.0
Terminalbench Hard(Stanford × Laude Institute (2026))
0.0
Tau Banking
0.0

LLM Stats Category Scores

(LLM Stats (zeroeval))
Structured Output
90
Instruction Following
90
Math
80
Legal
70
Multimodal
70
Image To Text
70
Finance
70
Grounding
70
Healthcare
70
Vision
70
Reasoning
60
Language
60
General
60
Code
60
Physics
40
Factuality
40
Biology
40
Chemistry
40
Long Context
10

Pricing

Input PriceFree
Output PriceFree
Blended Price (3:1)Free
Cache Read Price$0.04 / 1M tokens

Speed

Tokens/sec0.0
Time to First Token0.00s
Time to Answer0.00s

Provider Price Ranking

Provider Price Ranking

6 providers

Cheapest: OpenRouterMost Expensive: STACKIT
ProviderInputOutput
1OpenRouterCheapest
$0.08
$0.45
2Kilo Gateway
$0.08
$0.16
3Nebius Token Factory
$0.1
$0.3
4LLM Gateway
$0.1
$0.3
5NovitaAI
$0.119
$0.2
6STACKIT
$0.53
$0.76

Compare pricing across different API providers for this model.

External Sources