4-bit Open on Hugging Face
Parameters
107B
Weights (est.)
≈ 60.1 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/INTELLECT-3-4bit --prompt "Hello"
Details
- Base model
- PrimeIntellect/INTELLECT-3 (quantized)
- Vendor
- PrimeIntellect
- Task
- Text generation
- Quantisation
- 4-bit
- Licence
- mit
- Downloads
- 123 in the last 30 days · 1.1K all time · 3 likes
- Published
- 2025-11-27 · updated 2025-11-27 (11mo ago)
- Languages
- en
- Tags
- glm4_moe, prime-rl, verifiers, prime-intellect, reinforcement-learning, reasoning, agentic, mixture-of-experts, custom_code
Other MLX builds of PrimeIntellect/INTELLECT-3
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| INTELLECT-3-4bit (this one) | 4-bit | ≈ 60.1 GB | 96 GB+ | 123 |
| INTELLECT-3-5bit | 5-bit | ≈ 73.5 GB | 96 GB+ | 29 |
| INTELLECT-3-6bit | 6-bit | ≈ 86.8 GB | 128 GB+ | 30 |
| INTELLECT-3-8bit | 8-bit | ≈ 114 GB | 192 GB+ | 26 |
| INTELLECT-3-bf16 | bf16 | ≈ 214 GB | 256 GB+ | 30 |