Gemma 4 26B A4B
google/gemma-4-26b-a4b- Family · Gemma 4
- Released 2 Apr 2026
Not on OSMEXO yet
Scores · standing
Intelligence16.7#164/490
Coding39.3#92/196
News & trends
4 mentions this weekthis model · videos, posts, articles · 30 days
Trending · 14 days
- Gemma 4 26B-A4B running on an iPhone at 6–10 tok/s, in about 2 GB of RAM16 upvotes
- Gemma 4 26B-A4B on a 32GB Mac mini: Why It Feels Fast172 views
- 8 GB vs 24 GB GPU: Nemotron 3.5 Bolt E4B vs Gemma 4 26B A4B (Q4) #Shorts678 views
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#37/101138 tok/s1.2× faster than median
First token#190.5 s2.0× quicker than median
Family progress
Intelligence Index
- Family releases14.2
Research papers
- Apriel Voice 27B Preview: An Open Model for Speech, Reasoning, and Tool Use29 Oct
- The Router Already Knows20 Oct
- Towards Architectural-Level Safety Control: Relocating Hazardous Knowledge into Removable Experts in Mixture-of-Experts Models19 Oct
- The Concrete-Arbitrary Gap: Kinship Reasoning in LLMs Is Not Indifferent to Presentation30 Sep
- When Load-Balancing Goes Too Far: Expert Pruning in Over-Dispersed Mixture-of-Experts Models3 Sep
- RotaryQuant: Fitting 120B MoE Models on Consumer Hardware via Fused Compressed-Space Attention8 Aug
- Shape Mutating Expert Compression:LorExperts and BTExperts7 Aug
- Half the Experts, All the Code: One-Shot Domain Pruning of Mixture-of-Experts LLMs for Coding18 Jul
Social
Most watched and discussed · 90 days
- 8 GB vs 24 GB GPU: Nemotron 3.5 Bolt E4B vs Gemma 4 26B A4B (Q4) #ShortslocalLLMrig · 678 views · 26 Oct
- Gemma 4 26B-A4B on a 32GB Mac mini: Why It Feels FastPractical AI · 172 views · 31 Oct
- Gemma-4-26B-A4B: ThinkingCap cuts thinking tokens by 39%Homelab AI · 34 views · 24 Oct
- Gemma 4 26B-A4B QAT | AI Benchmark | UNIMATRIx Standard v2.1Gioele Stefano Luca Fierro · 33 views · 20 Oct
Same family
Gemma 4- 8.9
- 14.7
- 14.2
- 7.8
Alternatives
other vendors · similar capability- 12.9
- 18.2
- 21.5
- S19.5
Want Gemma 4 26B A4B on OSMEXO?
Sign in to request it. Requests are reviewed; a request does not guarantee availability.