Theia 21B V2A needs about 15.2 GB of memory at Q4_K_M (a 12.4 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 24 quants, from 7.2 GB (IQ2_M) to 40.9 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 9.4 GB.
Theia 21B V2A – a 20.4B llama model fine‑tuned for role‑play and storytelling
Theia 21B V2A is a 20.4B parameter llama model derived from TheDrummer/Theia-21B-v2. It supports ChatML, Alpaca, and Mistral formats for text completion and instruct tasks. The model is available in multiple GGUF quantizations.
Good at
- Creative storytelling
- Role‑play completion
- ChatML format support
Limits
- No vision support
- Large context requires high VRAM
- License type is other
Best for
- RP writers needing rich prose
- ChatML based chatbots
- Instruct‑guided story generation
Who it’s for. Developers and hobbyists who want a large llama model for text generation and role‑play. The model is suitable for local inference using llama.cpp or LM Studio.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ2_M | 7.2 GB | 10.0 GB | Too big | Get file → | |
| Q2_K | 7.8 GB | 10.6 GB | Too big | Get file → | |
| Q2_K_L | 8.5 GB | 11.3 GB | Too big | Get file → | |
| IQ3_XS | 8.6 GB | 11.5 GB | Too big | Get file → | |
| Q3_K_S | 9.0 GB | 11.9 GB | Too big | Get file → | |
| IQ3_M | 9.4 GB | 12.2 GB | Too big | Get file → | |
| Q3_K_M | 10.0 GB | 12.9 GB | Too big | Get file → | |
| Q3_K_L | 10.8 GB | 13.7 GB | Too big | Get file → | |
| IQ4_XS | 11.1 GB | 14.0 GB | Too big | Get file → | |
| Q3_K_XL | 11.4 GB | 14.3 GB | Too big | Get file → | |
| Q4_0 | 11.7 GB | 14.5 GB | Too big | Get file → | |
| Q4_0 | 11.7 GB | 14.5 GB | Too big | Get file → | |
| Q4_0 | 11.7 GB | 14.5 GB | Too big | Get file → | |
| Q4_0 | 11.7 GB | 14.6 GB | Too big | Get file → | |
| Q4_K_S | 11.7 GB | 14.6 GB | Too big | Get file → | |
| Q4_K_M | 12.4 GB | 15.2 GB | Too big | Get file → | |
| Q4_K_L | 12.9 GB | 15.7 GB | Too big | Get file → | |
| Q5_K_S | 14.1 GB | 17.0 GB | Too big | Get file → | |
| Q5_K_M | 14.5 GB | 17.3 GB | Too big | Get file → | |
| Q5_K_L | 14.9 GB | 17.8 GB | Too big | Get file → | |
| Q6_K | 16.8 GB | 19.6 GB | Too big | Get file → | |
| Q6_K_L | 17.1 GB | 19.9 GB | Too big | Get file → | |
| Q8_0 | 21.7 GB | 24.6 GB | Too big | Get file → | |
| F16 | 40.9 GB | 43.7 GB | Too big | Get file → |
For 8 GB VRAM use IQ2_M (7.19 GB) or Q2_K (7.8 GB). For 12 GB VRAM use Q3_K_M (10.01 GB). For 24 GB VRAM use Q8_0 (21.71 GB). Memory estimates are approximate.
What GPU do you need for Theia 21B V2A?
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 Q3_K_S (9.0 GB file, about 11.9 GB in use).
- 16 GB VRAM: the best quant that fits is Q4_K_L (12.9 GB file, about 15.7 GB in use).
- 24 GB VRAM: the best quant that fits is Q6_K_L (17.1 GB file, about 19.9 GB in use).
- 32 GB VRAM: the best quant that fits is Q8_0 (21.7 GB file, about 24.6 GB in use).
- 48 GB VRAM: the best quant that fits is F16 (40.9 GB file, about 43.7 GB in use).
Which Theia 21B V2A quant should you choose?
Q4_K_M is the usual starting point for Theia 21B V2A: it balances file size and quality well. Here it is a 12.4 GB file.
The Q8_0 file is 21.7 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 IQ2_M at 7.2 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 Theia 21B V2A
How much VRAM does Theia 21B V2A need?
About 15.2 GB at Q4_K_M with an 8K context window. That is the 12.4 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 Theia 21B V2A on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 10.0 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Theia 21B V2A GGUF?
It depends on the quant: IQ2_M is 7.2 GB, Q2_K is 7.8 GB, Q2_K_L is 8.5 GB, IQ3_XS is 8.6 GB, Q3_K_S is 9.0 GB, IQ3_M is 9.4 GB and 18 more.
What context length does Theia 21B V2A support?
The GGUF header lists a trained context length of 1,024,000 tokens (1000K). The KV cache grows with the context you use: about 2.3 GB at 8K and 9.4 GB at 32K.
Where can I download Theia 21B V2A GGUF?
The files are on Hugging Face in the bartowski/Theia-21B-v2-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 Theia 21B V2A 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 Theia 21B V2A on 8 GB VRAM?
Yes, use the IQ2_M or Q2_K quantizations which are 7.19 GB and 7.8 GB respectively.
Does the model support vision tasks?
No, the model has no vision capabilities.
What prompt formats are supported?
ChatML, Alpaca, and Mistral formats are supported for completion and instruct tasks.