Gemma 4 31B (Reasoning)
GoogleGemmaOpen WeightApache 2.0 · Commercial OK
Description
Gemma 4 31B is Google DeepMind's flagship dense multimodal model with 31 billion parameters and a 256K context window. Ranks #3 among open models on Arena AI. Built from the same research as Gemini 3, it features Per-Layer Embeddings, Shared KV Cache, alternating sliding-window and global attention, and variable aspect ratio vision encoding. Achieves an estimated LMArena text score of 1452.
Release Date
2026-04-02
Parameters
30.7B
Context Length
262K
Modalities
audio, image, text, video
Capability Radar
17
general
44
coding
86
reasoning
59
science
90
agents
70
multimodal
Rankings
| Domain | #Rank | Score | Source |
|---|---|---|---|
| Agentic Capability | 23 | 58.0 | LS |
| Code Ranking | 241 | 60.0 | AA |
| General Ranking | 249 | 48.0 | AA |
| Multimodal Ranking | 65 | 54.0 | LS |
| Science | 173 | 63.0 | AA |
Benchmark Scores (LLM Stats)
(LLM Stats (zeroeval))Agents
t2-bench
86.4%SR
Healthcare
MedXpertQA
61.3%SR
Language
MMMLU
88.4%SR
MMLU-Pro
85.2%SR
Long Context
MRCR v2 (8-needle)
66.4%SR
Math
AIME 2026
89.2%SR
MathVision
85.6%SR
Reasoning
GPQANYU + Cohere + Anthropic (2023)
84.3%SR
LiveCodeBench v6
80.0%SR
BIG-Bench Extra Hard
74.4%SR
Humanity's Last Exam
26.5%SR
Vision
MMMU-Pro
76.9%SR
AA Evaluation Indices
(Artificial Analysis)Gpqa(NYU + Cohere + Anthropic (2023))85.7
Ifbench(Google Research (2023))75.6
Lcr(Artificial Analysis)69.7
Tau2(Sierra + U Toronto + Vector Institute (2025))59.9
Scicode(UIUC + Argonne National Lab (2024))45.5
Terminalbench V2 143.4
Coding Index(Artificial Analysis)43.4
Terminalbench Hard(Stanford × Laude Institute (2026))36.4
Hle(Center for AI Safety + Scale AI (2025))23.6
Tau Banking14.8
Intelligence Index(Artificial Analysis)14.7
Terminalbench V4 00.0
LLM Stats Category Scores
(LLM Stats (zeroeval))Legal90
Finance90
Agents90
Tool Calling90
Language80
Physics80
Biology80
Chemistry80
Math70
Multimodal70
Reasoning70
General70
Long Context60
Healthcare60
Vision60
Pricing
Input PriceFree
Output PriceFree
Blended Price (3:1)Free
Speed
Tokens/sec35.0
Time to First Token0.98s
Time to Answer50.59s
Provider Price Ranking
Provider Price Ranking
21 providers
Cheapest: DeepInfraMost Expensive: Opper
ProviderInputOutput
1DeepInfraCheapest
$0
$0
2Novita
$0
$0
3FriendliAI
$0
$0
4Together
$0
$0
5OpenRouter
$0.09
$0.34
6Kilo Gateway
$0.09
$0.34
7NanoGPT
$0.1
$0.45
8DevPass (LLM Gateway)
$0.1
$0.25
9CrofAI
$0.1
$0.3
10Lilac
$0.11
$0.35
11FastRouter
$0.13
$0.38
12OrcaRouter
$0.13
$0.38
13Abacus
$0.14
$0.4
14NovitaAI
$0.14
$0.4
15Vercel AI Gateway
$0.14
$0.4
16Merge Gateway
$0.14
$0.4
17Crusoe
$0.14
$0.4
18Neuralwatt
$0.144
$0.42
19ai&
$0.2
$0.5
20Cortecs
$0.223
$0.39
21Opper
$0.46488
$2.44062
Compare pricing across different API providers for this model.