ruadapt qwen2.5 3B ext u48 instruct v4 needs about 2.7 GB of memory at Q4_K_M (a 1.9 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 9 quants, from 1.3 GB (Q2_K) to 6.2 GB. Longer context needs more: at 32K tokens the KV cache alone adds about 1.2 GB.
ruadapt qwen2.5 3B ext u48 instruct v4 – Russian‑adapted Qwen2.5 3B
This model is a fine‑tuned version of RefalMachine/ruadapt_qwen2.5_3B_ext_u48_full_lr5e4_peft_mlp_32_32_bs256. It uses an extended tiktoken cl100k tokenizer with 48k tokens and a learned embedding propagation technique. The model is designed for Russian language instruction tasks.
Good at
- Russian instruction completion
- Tool usage
- Fast generation on Russian text
Limits
- No vision support
- No license specified
- Context limited to 32768 tokens
Best for
- Chatbot in Russian
- Russian content generation
- Tool‑enabled assistants
Who it’s for. Developers building Russian language applications that require instruction following and tool integration. Suitable for teams with access to a GPU and sufficient VRAM to load the chosen quantization.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 1.3 GB | 2.1 GB | Fits | Get file → | |
| Q3_K_M | 1.6 GB | 2.4 GB | Fits | Get file → | |
| Q4_0 | 1.8 GB | 2.6 GB | Fits | Get file → | |
| Q4_K_M | 1.9 GB | 2.7 GB | Fits | Get file → | |
| Q5_0 | 2.2 GB | 3.0 GB | Fits | Get file → | |
| Q5_K_M | 2.2 GB | 3.0 GB | Fits | Get file → | |
| Q6_K | 2.5 GB | 3.3 GB | Fits | Get file → | |
| Q8_0 | 3.3 GB | 4.1 GB | Fits | Get file → | |
| — | 6.2 GB | 7.0 GB | Fits | Get file → |
For 8 GB VRAM, use Q4_0 (1.82 GB file, ~2.6 GB memory). For 12 GB VRAM, Q5_0 (2.16 GB file, ~3 GB memory) is a good balance. If more VRAM is available, Q6_K (2.53 GB file, ~3.3 GB memory) offers higher quality.
What GPU do you need for ruadapt qwen2.5 3B ext u48 instruct v4?
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: the best quant that fits is Q6_K (2.5 GB file, about 3.3 GB in use).
- 6 GB VRAM: the best quant that fits is Q8_0 (3.3 GB file, about 4.1 GB in use).
- 8 GB VRAM: the best quant that fits is the main file (6.2 GB file, about 7.0 GB in use).
Which ruadapt qwen2.5 3B ext u48 instruct v4 quant should you choose?
Q4_K_M is the usual starting point for ruadapt qwen2.5 3B ext u48 instruct v4: it balances file size and quality well. Here it is a 1.9 GB file.
The Q8_0 file is 3.3 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 1.3 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 ruadapt qwen2.5 3B ext u48 instruct v4
How much VRAM does ruadapt qwen2.5 3B ext u48 instruct v4 need?
About 2.7 GB at Q4_K_M with an 8K context window. That is the 1.9 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 ruadapt qwen2.5 3B ext u48 instruct v4 on an 8 GB GPU?
Yes, with the smaller files: the 6.2 GB file needs about 7.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of ruadapt qwen2.5 3B ext u48 instruct v4 GGUF?
It depends on the quant: Q2_K is 1.3 GB, Q3_K_M is 1.6 GB, Q4_0 is 1.8 GB, Q4_K_M is 1.9 GB, Q5_0 is 2.2 GB, Q5_K_M is 2.2 GB and 3 more.
What context length does ruadapt qwen2.5 3B ext u48 instruct v4 support?
The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 0.3 GB at 8K and 1.2 GB at 32K.
Where can I download ruadapt qwen2.5 3B ext u48 instruct v4 GGUF?
The files are on Hugging Face in the RefalMachine/ruadapt_qwen2.5_3B_ext_u48_instruct_v4_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?
Yes, the Q4_0 quantization requires about 2.6 GB VRAM at 8 k context, which fits within 8 GB.
Does the model support vision tasks?
No, the model does not include vision capabilities.
What context length can it handle?
It supports up to 32768 tokens of context.