LosslessMegaCoder Falcon 40b Mini needs about 26.7 GB of memory at Q4_K_M (a 25.2 GB file) with an 8K context window, so it fits on a 32 GB GPU. This repository has 22 quants, from 15.6 GB (Q2_K) to 43.9 GB (Q8_0).
LosslessMegaCoder Falcon 40b Mini is a 41.3B-parameter Falcon model in GGUF format, quantized by RichardErkhov. Its trained context length is 2K tokens. The repository has 22 quant files, from 15.6 GB (Q2_K) to 43.9 GB (Q8_0).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 15.6 GB | 16.3 GB | Too big | Get file → | |
| Q3_K_L | 16.5 GB | 17.2 GB | Too big | Get file → | |
| IQ3_XS | 17.5 GB | 18.3 GB | Too big | Get file → | |
| IQ3_S | 18.1 GB | 18.8 GB | Too big | Get file → | |
| Q3_K_S | 18.1 GB | 18.8 GB | Too big | Get file → | |
| IQ3_M | 18.9 GB | 19.7 GB | Too big | Get file → | |
| Q3_K | 19.9 GB | 20.6 GB | Too big | Get file → | |
| Q3_K_M | 19.9 GB | 20.6 GB | Too big | Get file → | |
| Q5_K_S | 22.5 GB | 23.3 GB | Too big | Get file → | |
| IQ4_XS | 22.5 GB | 23.3 GB | Too big | Get file → | |
| Q4_0 | 23.5 GB | 24.3 GB | Too big | Get file → | |
| Q4_K_S | 23.5 GB | 24.3 GB | Too big | Get file → | |
| IQ4_NL | 23.7 GB | 24.5 GB | Too big | Get file → | |
| Q4_K | 25.2 GB | 25.9 GB | Too big | Get file → | |
| Q4_K_M | 25.2 GB | 25.9 GB | Too big | Get file → | |
| Q4_1 | 26.1 GB | 26.8 GB | Too big | Get file → | |
| Q5_0 | 28.6 GB | 29.4 GB | Too big | Get file → | |
| Q5_K | 30.3 GB | 31.0 GB | Too big | Get file → | |
| Q5_K_M | 30.3 GB | 31.0 GB | Too big | Get file → | |
| Q5_1 | 31.2 GB | 31.9 GB | Too big | Get file → | |
| Q6_K | 34.0 GB | 34.8 GB | Too big | Get file → | |
| Q8_0 | 43.9 GB (2 parts) | 44.6 GB | Too big | Get file → |
What GPU do you need for LosslessMegaCoder Falcon 40b Mini?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: no quant fits at an 8K context.
- 16 GB VRAM: no quant fits at an 8K context.
- 24 GB VRAM: the best quant that fits is IQ4_XS (22.5 GB file, about 24.0 GB in use).
- 32 GB VRAM: the best quant that fits is Q5_K (30.3 GB file, about 31.8 GB in use).
- 48 GB VRAM: the best quant that fits is Q8_0 (43.9 GB file, about 45.4 GB in use).
Which LosslessMegaCoder Falcon 40b Mini quant should you choose?
Q4_K_M is the usual starting point for LosslessMegaCoder Falcon 40b Mini: it balances file size and quality well. Here it is a 25.2 GB file.
The Q8_0 file is 43.9 GB, about 1.7 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
If memory is tight, the smallest file is Q2_K at 15.6 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Frequently asked questions about LosslessMegaCoder Falcon 40b Mini
How much VRAM does LosslessMegaCoder Falcon 40b Mini need?
About 26.7 GB at Q4_K_M with an 8K context window. That is the 25.2 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run LosslessMegaCoder Falcon 40b Mini on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 17.1 GB, so you would need a larger GPU or CPU offloading.
What is the file size of LosslessMegaCoder Falcon 40b Mini GGUF?
It depends on the quant: Q2_K is 15.6 GB, Q3_K_L is 16.5 GB, IQ3_XS is 17.5 GB, IQ3_S is 18.1 GB, Q3_K_S is 18.1 GB, IQ3_M is 18.9 GB and 16 more.
What context length does LosslessMegaCoder Falcon 40b Mini support?
The GGUF header lists a trained context length of 2,048 tokens (2K). The KV cache grows with the context you use: about 1.0 GB at 8K and 0.3 GB at 2K.
Where can I download LosslessMegaCoder Falcon 40b Mini GGUF?
The files are on Hugging Face in the RichardErkhov/rombodawg_-_LosslessMegaCoder-Falcon-40b-mini-gguf repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does LosslessMegaCoder Falcon 40b Mini use?
Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.