MI

Codestral-22B-v0.1

MistralGGUF by duyntnetUpdated 2024-06-23
Download on Hugging Face
Parameters22.2B
Context32K
Layers56
Architecturellama
Smallest quant4.8 GB
Quick answer

Codestral-22B-v0.1 needs about 15.7 GB of memory at Q4_K_M (a 13.3 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 27 quants, from 4.8 GB (IQ1_S) to 23.6 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 7.5 GB.

22B parameter LLM focused on coding

Codestral-22B-v0.1 is a 22.2B parameter model in the Mistral family. It is trained on a 32768 token context window and is optimized for code generation tasks.

Good at

  • Code generation
  • Programming assistance
  • Syntax completion

Limits

  • No vision support
  • License type not specified
  • Requires large VRAM for higher precision

Best for

  • Writing functions in Rust
  • Debugging code snippets
  • Generating API documentation

Who it’s for. Developers and programmers who need a large language model specialized in coding tasks. The model is suitable for integration into coding assistants and IDE plugins.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_S4.8 GB 7.2 GB
Tight Get file →
IQ1_M5.3 GB 7.6 GB
Tight Get file →
IQ2_XXS6.0 GB 8.4 GB
Too big Get file →
IQ2_XS6.6 GB 9.0 GB
Too big Get file →
IQ2_S7.0 GB 9.4 GB
Too big Get file →
IQ2_M7.6 GB 10.0 GB
Too big Get file →
Q2_K_S7.7 GB 10.0 GB
Too big Get file →
Q2_K8.3 GB 10.7 GB
Too big Get file →
IQ3_XXS8.6 GB 11.0 GB
Too big Get file →
IQ3_XS9.2 GB 11.6 GB
Too big Get file →
Q3_K_S9.6 GB 12.0 GB
Too big Get file →
IQ3_S9.7 GB 12.1 GB
Too big Get file →
IQ3_M10.1 GB 12.4 GB
Too big Get file →
Q3_K_M10.8 GB 13.1 GB
Too big Get file →
Q3_K_L11.7 GB 14.1 GB
Too big Get file →
IQ4_XS11.9 GB 14.3 GB
Too big Get file →
IQ4_NL12.6 GB 15.0 GB
Too big Get file →
Q4_012.6 GB 15.0 GB
Too big Get file →
Q4_K_S12.7 GB 15.0 GB
Too big Get file →
Q4_K_M13.3 GB 15.7 GB
Too big Get file →
Q4_113.9 GB 16.3 GB
Too big Get file →
Q5_K_S15.3 GB 17.7 GB
Too big Get file →
Q5_015.4 GB 17.7 GB
Too big Get file →
Q5_K_M15.7 GB 18.1 GB
Too big Get file →
Q5_116.7 GB 19.1 GB
Too big Get file →
Q6_K18.3 GB 20.6 GB
Too big Get file →
Q8_023.6 GB 26.0 GB
Too big Get file →

For 8 GB VRAM use IQ1_S or IQ1_M. For 16 GB VRAM use Q2_K or Q3_K_M. For 32 GB VRAM use Q4_K_M or higher. Memory estimates are approximate.

What GPU do you need for Codestral-22B-v0.1?

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: the best quant that fits is IQ1_M (5.3 GB file, about 7.6 GB in use).
  • 12 GB VRAM: the best quant that fits is Q3_K_S (9.6 GB file, about 12.0 GB in use).
  • 16 GB VRAM: the best quant that fits is Q4_K_M (13.3 GB file, about 15.7 GB in use).
  • 24 GB VRAM: the best quant that fits is Q6_K (18.3 GB file, about 20.6 GB in use).
  • 32 GB VRAM: the best quant that fits is Q8_0 (23.6 GB file, about 26.0 GB in use).

Which Codestral-22B-v0.1 quant should you choose?

Q4_K_M is the usual starting point for Codestral-22B-v0.1: it balances file size and quality well. Here it is a 13.3 GB file.

The Q8_0 file is 23.6 GB, about 1.8 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 IQ1_S at 4.8 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

Architecturellama
Parameters22.2B
Layers56
Trained context32,768 tokens
KV cache (fp16)224 KB per token

Source

Quantized byduyntnet
FamilyMistral
Licenseother
Downloads (HF)606
Last updated2024-06-23

Frequently asked questions about Codestral-22B-v0.1

How much VRAM does Codestral-22B-v0.1 need?

About 15.7 GB at Q4_K_M with an 8K context window. That is the 13.3 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-v0.1 on an 8 GB GPU?

Yes, with IQ1_M: the 5.3 GB file needs about 7.6 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Codestral-22B-v0.1 GGUF?

It depends on the quant: IQ1_S is 4.8 GB, IQ1_M is 5.3 GB, IQ2_XXS is 6.0 GB, IQ2_XS is 6.6 GB, IQ2_S is 7.0 GB, IQ2_M is 7.6 GB and 21 more.

What context length does Codestral-22B-v0.1 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-v0.1 GGUF?

The files are on Hugging Face in the duyntnet/Codestral-22B-v0.1-imatrix-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-v0.1 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-v0.1 on 8 GB VRAM?

Use the IQ1_S or IQ1_M quantization; they require about 7.2–7.6 GB of VRAM at 8k context.

Does this model support image inputs?

No, the model has no vision capabilities.

What context length does it support?

It supports up to 32768 tokens.

Similar models

Browse all →

More from duyntnet

All by duyntnet →

More Mistral models

All Mistral →

Other medium models

All medium models →