bf16 Open on Hugging Face
Parameters
7.8B
Weights (est.)
≈ 15.6 GB
Mac memory (est.)
24 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/EXAONE-Deep-7.8B-bf16 --prompt "Hello"
Details
- Base model
- LGAI-EXAONE/EXAONE-Deep-7.8B (finetune)
- Vendor
- LGAI-EXAONE
- Task
- Text generation
- Quantisation
- bf16
- Licence
- exaone
- Downloads
- 31 in the last 30 days · 212 all time · 0 likes
- Published
- 2025-03-18 · updated 2025-03-18 (1y ago)
- Languages
- en, ko
- Tags
- exaone, lg-ai, exaone-deep, custom_code
Other MLX builds of LGAI-EXAONE/EXAONE-Deep-7.8B
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| EXAONE-Deep-7.8B-3bit | 3-bit | ≈ 3.4 GB | 8 GB+ | 24 |
| EXAONE-Deep-7.8B-4bit | 4-bit | ≈ 4.4 GB | 8 GB+ | 49 |
| EXAONE-Deep-7.8B-6bit | 6-bit | ≈ 6.4 GB | 16 GB+ | 20 |
| EXAONE-Deep-7.8B-8bit | 8-bit | ≈ 8.3 GB | 16 GB+ | 23 |
| EXAONE-Deep-7.8B-bf16 (this one) | bf16 | ≈ 15.6 GB | 24 GB+ | 31 |
| EXAONE-Deep-7.8B-fp16 | fp16 | ≈ 15.6 GB | 24 GB+ | 25 |