Gemma 3 27B Instruct
GoogleGemmaOpen WeightGemma · Usage Commercial
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.
Date de sortie
2025-03-12
Paramètres
27.0B
Longueur du contexte
131K
Modalités
image, text
Radar de capacités
23
general
13
coding
34
reasoning
28
science
28
agents
70
multimodal
Classements
| Domaine | #Rang | Score | Source |
|---|---|---|---|
| Classement codage | 575 | 11.0 | AA |
| Classement général | 546 | 24.0 | AA |
| Classement multimodal | 109 | 41.0 | LS |
| Science | 547 | 23.0 | AA |
Scores de benchmarks (LLM Stats)
(LLM Stats (zeroeval))Chat
IFEvalGoogle Research (2023)
90.4%Aut.
Factuality
SimpleQA
10.0%Aut.
General
Global-MMLU-Lite
75.1%Aut.
Language
MMLU-Pro
67.5%Aut.
WMT24++
53.4%Aut.
ECLeKTic
16.7%Aut.
Long Context
MRCR v2 (8-needle)
13.5%Aut.
Math
GSM8k
95.9%Aut.
MATH
89.0%Aut.
MathVista-Mini
67.6%Aut.
HiddenMath
60.3%Aut.
Reasoning
HumanEvalOpenAI (2021)
87.8%Aut.
BIG-Bench Hard
87.6%Aut.
Natural2Code
84.5%Aut.
ChartQAMasry et al. (2022)
78.0%Aut.
FACTS Grounding
74.9%Aut.
MBPP
0.74 / 100Aut.
Bird-SQL (dev)
54.4%Aut.
GPQANYU + Cohere + Anthropic (2023)
42.4%Aut.
LiveCodeBench
29.7%Aut.
BIG-Bench Extra Hard
19.3%Aut.
Vision
DocVQADocVQA (2020)
86.6%Aut.
AI2D
84.5%Aut.
VQAv2 (val)
71.0%Aut.
InfoVQA
70.6%Aut.
TextVQA
65.1%Aut.
MMMU (val)
64.9%Aut.
Indices d'évaluation AA
(Artificial Analysis)Math 500(OpenAI (2024), subset of Hendrycks et al. MATH (2021))88.3
Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))66.9
Gpqa(NYU + Cohere + Anthropic (2023))42.8
Ifbench(Google Research (2023))31.8
Aime(MAA (Mathematical Association of America))25.3
Scicode(UIUC + Argonne National Lab (2024))23.3
Math Index(Artificial Analysis)20.7
Aime 25(MAA (Mathematical Association of America))20.7
Livecodebench(UC Berkeley + MIT + Cornell (2024))13.7
Tau2(Sierra + U Toronto + Vector Institute (2025))10.5
Coding Index(Artificial Analysis)10.1
Lcr(Artificial Analysis)7.3
Intelligence Index(Artificial Analysis)4.9
Terminalbench V2 14.5
Hle(Center for AI Safety + Scale AI (2025))4.4
Terminalbench Hard(Stanford × Laude Institute (2026))3.8
Tau Banking0.8
Terminalbench V4 00.0
Scores par catégorie LLM Stats
(LLM Stats (zeroeval))Chat90
Instruction Following90
Structured Output90
Math80
Image To Text70
Legal70
Multimodal70
Finance70
Grounding70
Healthcare70
Vision70
Language60
Reasoning60
General60
Code60
Physics40
Factuality40
Biology40
Chemistry40
Long Context10
Tarification
Prix d'entréeGratuit
Prix de sortieGratuit
Prix mixte (3:1)Gratuit
Prix de lecture cache$0.04 / 1M tokens
Vitesse
Tokens/sec0.0
Délai du premier token0.00s
Temps de réponse0.00s
Classement des Prix par Fournisseur
Classement des Prix par Fournisseur
9 fournisseurs
Moins cher: DeepInfraPlus cher: STACKIT
FournisseurEntréeSortie
1DeepInfraMoins cher
$0
$0
2OpenRouter
$0.08
$0.45
3Hugging Face
$0.08
$0.16
4Deep Infra
$0.08
$0.16
5Kilo Gateway
$0.08
$0.16
6Merge Gateway
$0.08
$0.45
7Nebius Token Factory
$0.1
$0.3
8NovitaAI
$0.119
$0.2
9STACKIT
$0.53
$0.76
Comparer les prix entre différents fournisseurs API pour ce modèle.