koishi 8x7b qlora i1 needs about 30.0 GB of memory at Q4_K_M (a 28.4 GB file) with an 8K context window, so it fits on a 32 GB GPU. This repository has 21 quants, from 9.8 GB (IQ1_S) to 38.4 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.
Koishi 8x7b QLoRA i1 – 46.7B LLaMA‑style model
Koishi 8x7b QLoRA i1 is a 46.7B parameter LLaMA‑based model trained with QLoRA on a 4x NVIDIA A100 cluster. It supports a 32768 token context window and is available in multiple GGUF quantizations.
Good at
- Large context reasoning
- Fine‑tuned role prompts
- High‑capacity language modeling
Limits
- No vision support
- License not specified
- Requires large VRAM for full precision
Best for
- Chatbot development
- Long‑form content generation
- Research prototyping
Who it’s for. Developers and researchers needing a high‑capacity LLaMA model with extended context. Suitable for environments with sufficient GPU memory to load the chosen quantization.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ1_S | 9.8 GB | 11.4 GB | Too big | Get file → | |
| IQ1_M | 10.8 GB | 12.4 GB | Too big | Get file → | |
| IQ2_XXS | 12.6 GB | 14.1 GB | Too big | Get file → | |
| IQ2_XS | 13.9 GB | 15.5 GB | Too big | Get file → | |
| IQ2_S | 14.1 GB | 15.7 GB | Too big | Get file → | |
| IQ2_M | 15.5 GB | 17.1 GB | Too big | Get file → | |
| Q2_K | 17.3 GB | 18.9 GB | Too big | Get file → | |
| IQ3_XXS | 18.2 GB | 19.8 GB | Too big | Get file → | |
| IQ3_XS | 19.4 GB | 20.9 GB | Too big | Get file → | |
| IQ3_S | 20.4 GB | 22.0 GB | Too big | Get file → | |
| Q3_K_S | 20.4 GB | 22.0 GB | Too big | Get file → | |
| IQ3_M | 21.4 GB | 23.0 GB | Too big | Get file → | |
| Q3_K_M | 22.5 GB | 24.1 GB | Too big | Get file → | |
| Q3_K_L | 24.2 GB | 25.7 GB | Too big | Get file → | |
| IQ4_XS | 25.1 GB | 26.7 GB | Too big | Get file → | |
| Q4_0 | 26.6 GB | 28.1 GB | Too big | Get file → | |
| Q4_K_S | 26.7 GB | 28.3 GB | Too big | Get file → | |
| Q4_K_M | 28.4 GB | 30.0 GB | Too big | Get file → | |
| Q5_K_S | 32.2 GB | 33.8 GB | Too big | Get file → | |
| Q5_K_M | 33.2 GB | 34.8 GB | Too big | Get file → | |
| Q6_K | 38.4 GB | 40.0 GB | Too big | Get file → |
For 8 GB VRAM, use IQ1_S (file size 9.82 GB, estimated 11.4 GB at 8k context). For 12 GB VRAM, IQ1_M (10.85 GB, 12.4 GB). For 16 GB VRAM, IQ2_XS (13.92 GB, 15.5 GB). Higher quants require proportionally more memory.
What GPU do you need for koishi 8x7b qlora i1?
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: the best quant that fits is IQ1_S (9.8 GB file, about 11.4 GB in use).
- 16 GB VRAM: the best quant that fits is IQ2_S (14.1 GB file, about 15.7 GB in use).
- 24 GB VRAM: the best quant that fits is IQ3_M (21.4 GB file, about 23.0 GB in use).
- 32 GB VRAM: the best quant that fits is Q4_K_M (28.4 GB file, about 30.0 GB in use).
- 48 GB VRAM: the best quant that fits is Q6_K (38.4 GB file, about 40.0 GB in use).
Which koishi 8x7b qlora i1 quant should you choose?
Q4_K_M is the usual starting point for koishi 8x7b qlora i1: it balances file size and quality well. Here it is a 28.4 GB file.
If memory is tight, the smallest file is IQ1_S at 9.8 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
Source
Frequently asked questions about koishi 8x7b qlora i1
How much VRAM does koishi 8x7b qlora i1 need?
About 30.0 GB at Q4_K_M with an 8K context window. That is the 28.4 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 koishi 8x7b qlora i1 on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 11.4 GB, so you would need a larger GPU or CPU offloading.
What is the file size of koishi 8x7b qlora i1 GGUF?
It depends on the quant: IQ1_S is 9.8 GB, IQ1_M is 10.8 GB, IQ2_XXS is 12.6 GB, IQ2_XS is 13.9 GB, IQ2_S is 14.1 GB, IQ2_M is 15.5 GB and 15 more.
What context length does koishi 8x7b qlora i1 support?
The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 1.1 GB at 8K and 4.3 GB at 32K.
Where can I download koishi 8x7b qlora i1 GGUF?
The files are on Hugging Face in the mradermacher/koishi-8x7b-qlora-i1-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
Can I run this model on 8 GB VRAM?
Use the IQ1_S quant, which is 9.82 GB and estimates 11.4 GB at 8k context. It may fit with careful batch sizing.
Does this model support vision tasks?
No, the model has no vision capability.
What is the maximum context length?
The model supports 32768 tokens.