Gemma 4 31B
google/gemma-4-31b- Family · Gemma 4
- Released 2 Apr 2026
- Vision
Not on OSMEXO yet
Spec & list price
Context256K
Scores · standing
Intelligence14.7#178/490
LiveBench38.7#32/32
Coding43.4#84/196
News & trends
10 mentions this week ▲ 11%this model · videos, posts, articles · 30 days
Trending · 14 days
- Gemma-4-31B-Humanlike-Chat: 2.0's improved human voice on official Gemma 4 31B319 upvotes
- Qwen 3.8 27B vs Gemma 4 31B on RTX 5090: 4 Real Coding Tests2.3K views
- Qwen 3.8 27B vs Gemma 4 31B on RTX 5090: 4 Real Coding Tests7.6K views
- Gemini 3.6 Flash won a nuclear last stand without breaking a single pact. Google's open Gemma 4 31B came 2nd of 43.69 upvotes
News · latest from Google
- Gemini 4 Carbon is the new #1 on the Artificial Analysis Intelligence Index v529 Oct
- Gemini 4 Carbon (High) Scores 59 on Artificial Analysis v528 Oct
- Gemini 4 Carbonblog.google · 28 Oct
- Gemini 4 Carbon: Our next frontier modelGoogle DeepMind · 28 Oct
Benchmarks
bar = share of the top score · Benchmark data: Artificial AnalysisSpeed · median across providers, latest day
Output speed#81/10161 tok/s1.9× slower than median
First token#270.56 s1.7× quicker than median
Family progress
Intelligence Index
- Family releases14.2
LiveBench suites
Browser0.0
Coding49.0
Data Analysis53.3
Design58.4
Long Agentic10.7
Reasoning68.3
Research44.4
▲ Strongest: Reasoning, Design. ▼ Weakest: Browser (0).
Research papers
- Beyond Passing Tests: A Systematic Review and Controlled Re-evaluation of Techniques for Improving the Quality of LLM-Generated Code28 Oct
- The Lay of the Land in Idiom-Translation Quality Estimation26 Oct
- Many Perturbations, Shared Failure Dynamics: Tracing Direction-Sensitive Perturbation Influence in Language Models23 Oct
- Joules-to-GDP: How Much Economic Value Do AI Systems Produce per Unit of Energy?22 Oct
- Benchmarking Large Language Models for Detecting Satirical Code-Mixed Sentences21 Oct
- CHSM8: A Chess Transformer that Follows Natural-Language Style Prompts18 Oct
- BanglaShield: A Script-Robust Safety Classifier and Benchmark for Bangla Native, Romanized, and Code-Mixed Prompts16 Oct
- SolBehave: Evaluating the Behavioural Correctness of LLM-Generated Smart Contracts13 Oct
Social
Most watched and discussed · 90 days
- Qwen 3.8 27B vs Gemma 4 31B on RTX 5090: 4 Real Coding TestsLaxmi Kant | KGP Talkie · 7.6K views · 29 Oct
- Qwen 3.8 27B vs Gemma 4 31B on RTX 5090: 4 Real Coding TestsLaxmi Kant | KGP Talkie · 2.3K views · 29 Oct
- Gemma 4 31B vs Recursive Flint 40B: Live Coding Test & 24 GB VRAM Math #ShortslocalLLMrig · 370 views · 25 Oct
- Qwen 3.8 27B vs Gemma 4 31B on RTX 5090: 4 Real Coding Tests | Kaun Jeeta?Laxmi Kant - KGP Talkie Hindi · 300 views · 29 Oct
- Gemma 4 31B vs Qwen 3.6 35B — Live Side-by-Side Test (0 GB GPU Needed) #ShortslocalLLMrig · 119 views · 25 Oct
- NVIDIAのGemma 4 31B（4bit NVFP4）をRTX 5090一枚で動かしてみた｜6問は採用、でも重さ30GBで毎秒2語｜ローカルAI検証さつきのOSS研究室 · 108 views · 24 Oct
- GEMMA 4 31B LOCAL RUN: Run Google's 30B MoE Giant Locally! #Gemma4 #AI #TechSystem Stack · 95 views · 23 Oct
- Gemma 4 31B - Inside a TransformerThe Residual Stream · 23 views · 23 Oct
- GEMMA 4 31B: How to Run Google's Giant on Weak HardwareSystem Stack · 11 views · 31 Oct
- Claude Sonnet 5 vs Gemma 4 31B: 7-Round Sprint | Eigendark Model Arena 084Eigendark · 8 views · 31 Oct
- Gemma 4 31B: Run Google's Open Giant on Weak PCSystem Stack · 4 views · 1 Nov
Same family
Gemma 4- 8.9
- 14.2
- 16.7
- 7.8
Alternatives
other vendors · similar capability- 12.9
- 11.6
- 18.2
- 10.0
Want Gemma 4 31B on OSMEXO?
Sign in to request it. Requests are reviewed; a request does not guarantee availability.