Buddy26B
The brain of Buddy: Your AI Friend (iOS), by Buddy's Burrow. It runs entirely on the phone.
Built on Google's Gemma 4 26B-A4B (instruction-tuned), using Unsloth's UD-IQ2_M quantization (unsloth/gemma-4-26B-A4B-it-GGUF), split so the experts can be streamed from flash:
| File | What |
|---|---|
buddy26b.json |
Map of the split (which expert lives where) |
buddy26b.resident.gguf |
Always-used weights (5-bit), 1.78 GB |
buddy26b.experts |
Experts, read from flash as needed, 8.27 GB |
buddy26b-eyes.gguf |
Vision projector (ggml-org's Gemma 4 26B-A4B mmproj, Q8_0), 0.81 GB |
Gemma 4 is released by Google under the Apache License 2.0 (see google/gemma-4-26B-A4B-it); this split keeps that license.
- Downloads last month
- -
Hardware compatibility
Log In to add your hardware
We're not able to determine the quantization variants.
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support