4-bit Open on Hugging Face
Parameters
104B
Weights (est.)
≈ 58.6 GB
Mac memory (est.)
96 GB+
Size is estimated from the parameter count and quantisation. Memory allows a fifth extra for context plus 2 GB for the system; long contexts need more.
Run it
With mlx-lm on an Apple silicon Mac:
pip install mlx-lmmlx_lm.generate --model mlx-community/Ring-flash-linear-2.0-128k-4bit --prompt "Hello"
Details
- Base model
- inclusionAI/Ring-flash-linear-2.0-128k (quantized)
- Vendor
- inclusionAI
- Task
- Text generation
- Quantisation
- 4-bit
- Licence
- mit
- Downloads
- 35 in the last 30 days · 343 all time · 1 likes
- Published
- 2025-10-23 · updated 2025-10-23 (1y ago)
- Languages
- en
- Tags
- bailing_moe_linear, custom_code
Other MLX builds of inclusionAI/Ring-flash-linear-2.0-128k
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| Ring-flash-linear-2.0-128k-3bit | 3-bit | ≈ 45.6 GB | 64 GB+ | 23 |
| Ring-flash-linear-2.0-128k-4bit (this one) | 4-bit | ≈ 58.6 GB | 96 GB+ | 35 |
| Ring-flash-linear-2.0-128k-5bit | 5-bit | ≈ 71.6 GB | 96 GB+ | 24 |
| Ring-flash-linear-2.0-128k-6bit | 6-bit | ≈ 84.6 GB | 128 GB+ | 19 |
| Ring-flash-linear-2.0-128k-8bit | 8-bit | ≈ 111 GB | 128 GB+ | 20 |