Kimi-Linear-48B-A3B-Instruct
Kimi-Linear-48B-A3B-Instruct is an open-weight language model from Moonshot AI with 49.1B parameters (mixture of experts). At 4-bit it needs about 34.7 GB of memory, so the smallest current device that fits it is the L40S.
Overview
Includes room for an 8k context and runtime overhead. More context needs more memory.
Key information
- Lab
- Moonshot AI
- Released
- 30 Oct 2025
- Parameters
- 49.1B · MoE
- Type
- Language
- Context
- Unknown
- License
- MIT
- HF downloads, 30 days
- 44,661
- Repository
- moonshotai/Kimi-Linear-48B-A3B-Instruct
More from Moonshot AI
View allWill it run on your hardware?
How we estimate| Estimate or device | Memory |
|---|---|
| Q4 | 34.7 GB |
| Q8 | 59.5 GB |
| BF16 | 110.3 GB |
| RTX 5060 Ti 16 GB | 16 GB usable |
| RTX 5090 32 GB | 32 GB usable |
| DGX Spark 128 GB | 108.8 GB usable |
| Mac Studio M5 Ultra 256 GB | 192 GB usable |
Single-device memory estimate. Pick a device to update the On your device column.
36 GB usable memory
| Format | Memory needed | On your device | Example hardware |
|---|---|---|---|
| 4-bit (Q4) | ≈ 34.7 GB | Tight 96% | L40S48 GB memory |
| 8-bit (Q8) | ≈ 59.5 GB | Too big 165% | A100 80GB SXM80 GB memory |
| 16-bit (BF16) | ≈ 110.3 GB | Too big 306% | H200 SXM141 GB memory |
Fits means the estimate stays within 90% of usable memory. Tight is above 90% and up to 100%. Too big is above 100%. These are memory checks, not speed or compatibility tests.
Mac allowances are 66% of total RAM up to 36 GB, and 75% above. AI boxes allow 85% because their memory is shared with the operating system.
Where to run it
On a Mac
Start with the MLX files, built for Apple Silicon.
mlx-community/Kimi-Linear-48B-A3B-Instruct-4bitCheck your MacThrough an API
No hosted API found.
mlx_lm.generate --model mlx-community/Kimi-Linear-48B-A3B-Instruct-4bit \
--prompt "Explain how a transformer works."Example commands built from the repository ids above. They need a working install of MLX. We have not run them; check each repository for available files.
More about this model
Frequently asked questions
How much memory does Kimi-Linear-48B-A3B-Instruct need? At an 8k context, about 34.7 GB at 4-bit, 59.5 GB at 8-bit and 110.3 GB at 16-bit (BF16). A longer context needs more.
Can I run Kimi-Linear-48B-A3B-Instruct on a Mac? The smallest current Mac that fits the 4-bit estimate of 34.7 GB is the Mac mini M5 Pro (64GB). An MLX build is available: mlx-community/Kimi-Linear-48B-A3B-Instruct-4bit.
Does the full context fit? We do not have a context length for this model. Our memory estimates use 8k, so long prompts need more memory than shown.
What does it cost through an API? We did not find a hosted API for this model.
License in plain English
MIT. This license permits commercial use, modification and redistribution, as long as the copyright and license notice are kept. This summary is not legal advice.
OMHQ RigBench
Not yet tested. We show no speed or quality score without a measurement.
Revision history and sources
Last updated . Sources:Hugging Face model card, MLX repository.