Gemma 4 31B (Reasoning)
GoogleGemmaOpen WeightApache 2.0 · Commercial OK
描述
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.
發布日期
2026-04-02
參數規模
30.7B
上下文長度
262K
支援模態
image, text, video
能力雷達圖
36
general
39
coding
86
reasoning
58
science估算
90
agents
70
multimodal
Science 在缺少專門科學評測時使用推理能力代理估算。
排行榜排名
基準測試分數 (LLM Stats)
Agents
t2-bench
86.4%自報
Biology
GPQA
84.3%自報
Finance
MMLU-Pro
85.2%自報
General
MMMLU
88.4%自報
LiveCodeBench v6
80.0%自報
MMMU-Pro
76.9%自報
BIG-Bench Extra Hard
74.4%自報
MRCR v2
66.4%自報
Healthcare
MedXpertQA
61.3%自報
Math
AIME 2026
89.2%自報
MathVision
85.6%自報
Humanity's Last Exam
26.5%自報
AA 評測指數
Intelligence Index39.2
Coding Index38.7
Gpqa0.9
Ifbench0.8
Lcr0.6
Tau20.6
Scicode0.4
Terminalbench Hard0.4
Hle0.2
LLM Stats 分類評分
Tool Calling90
Agents90
Finance90
Legal90
Biology80
Chemistry80
General80
Language80
Physics80
Math70
Multimodal70
Reasoning70
Vision60
Healthcare60
Long Context60
定價
輸入價格免費
輸出價格免費
混合價格(3:1)免費
速度
Tokens/秒34.9 tokens/s
首Token延遲1.02s
首回答延遲58.33s
可用提供商
(LS 內部計價單位)| 提供商 | 輸入價格 | 輸出價格 |
|---|---|---|
| Novita | 140K | 400K |