Codestral 22B RAG Q8 needs about 26.0 GB of memory at its main file (a 23.6 GB file) with an 8K context window, so it fits on a 32 GB GPU. Longer context needs more: at 32K tokens the KV cache alone adds about 7.5 GB.
Codestral 22B RAG Q8 – a 22.2B parameter model focused on context‑obedient RAG.
Codestral 22B RAG Q8 is a 22.2B parameter model designed for retrieval‑augmented generation. It follows a strict prompt format to improve adherence to supplied context and reduce hallucinations. The model is tailored for coding and RAG tasks.
Good at
- Context‑obedient RAG
- Coding assistance
- Source‑aware responses
Limits
- Requires large VRAM – 26 GB at 8 k context
- No vision support
- License type is other
Best for
- RAG pipelines needing source citations
- Code generation with context references
- Academic research with strict context adherence
Who it’s for. Developers and researchers building retrieval‑augmented applications that need reliable source attribution and minimal hallucinations. The model is suitable for teams with access to high‑VRAM GPUs.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 23.6 GB | 26.0 GB | Too big | Get file → |
The only available quant is Q8. It requires approximately 26 GB of VRAM at 8 k context. For lower VRAM budgets, no other quant is provided, so the model cannot be run on GPUs with less than 26 GB of VRAM.
What GPU do you need for Codestral 22B RAG Q8?
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: no quant fits at an 8K context.
- 32 GB VRAM: the best quant that fits is the main file (23.6 GB file, about 26.0 GB in use).
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Codestral 22B RAG Q8
How much VRAM does Codestral 22B RAG Q8 need?
About 26.0 GB at its main file with an 8K context window. That is the 23.6 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 Codestral 22B RAG Q8 on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 26.0 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Codestral 22B RAG Q8 GGUF?
It depends on the quant: file is 23.6 GB.
What context length does Codestral 22B RAG Q8 support?
The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 1.9 GB at 8K and 7.5 GB at 32K.
Where can I download Codestral 22B RAG Q8 GGUF?
The files are on Hugging Face in the TroyDoesAI/Codestral-22B-RAG-Q8-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 Codestral 22B RAG Q8 use?
Hugging Face lists the license as other. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run Codestral 22B RAG Q8 on an RTX 3090 with 24 GB VRAM?
No. The model needs about 26 GB of VRAM at 8 k context, which exceeds the 24 GB limit of an RTX 3090.
Does this model support vision tasks?
No. The model is designed for text and coding tasks only.
What prompt format should I use?
Use the explicit delimiters: Contextual-Request, BEGININPUT, BEGINCONTEXT, ENDCONTEXT, ENDINPUT, BEGININSTRUCTION, ENDINSTRUCTION as described in the repo.