csm 1b needs about 2.7 GB of memory at Q8_0 (a 2.0 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 3 quants, from 1.4 GB (Q4_K) to 3.6 GB (F16).
CSM-1B – a 1.78B parameter TTS model in GGUF format
CSM-1B is a conversational speech model that converts text to 24 kHz PCM audio. It uses a two‑stage transformer architecture with a 16‑layer backbone and a 4‑layer depth decoder. The model is released under Apache‑2.0.
Good at
- Text‑to‑speech conversion
- 24 kHz audio output
- Two‑stage transformer decoding
Limits
- No vision support
- No explicit language list
- Requires 8 k context for memory estimates
Best for
- Embedded TTS applications
- Low‑latency speech synthesis
- Custom voice generation
Who it’s for. Developers building speech interfaces who need a lightweight, open‑source TTS model. The model is suitable for projects that can run GGUF files with llama.cpp or similar engines.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K | 1.4 GB | 2.1 GB | Fits | Get file → | |
| Q8_0 | 2.0 GB | 2.7 GB | Fits | Get file → | |
| F16 | 3.6 GB | 4.5 GB | Fits | Get file → |
For 8 GB VRAM, Q4_K (1.44 GB, 2.1 GB memory at 8 k context) is recommended. If more VRAM is available, Q8_0 (2.01 GB, 2.7 GB memory) offers higher precision. F16 (3.57 GB, 4.5 GB memory) provides full‑precision weights but needs larger memory.
What GPU do you need for csm 1b?
These figures add the file size, an fp16 KV cache at an 8K context window 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 Q8_0 (2.0 GB file, about 2.7 GB in use).
- 6 GB VRAM: the best quant that fits is F16 (3.6 GB file, about 4.5 GB in use).
Which csm 1b quant should you choose?
Q8_0 (2.0 GB) is the middle option in this repository. Close to lossless. It costs roughly twice the memory of a 4-bit file, so it suits cases where you have plenty of VRAM.
If memory is tight, the smallest file is Q4_K at 1.4 GB, with the quality trade-offs that come with fewer bits. The usual sweet spot between file size and quality, and the quant most people start with.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about csm 1b
How much VRAM does csm 1b need?
About 2.7 GB at Q8_0 with an 8K context window. That is the 2.0 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 csm 1b on an 8 GB GPU?
Yes, with F16: the 3.6 GB file needs about 4.5 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of csm 1b GGUF?
It depends on the quant: Q4_K is 1.4 GB, Q8_0 is 2.0 GB, F16 is 3.6 GB.
Where can I download csm 1b GGUF?
The files are on Hugging Face in the cstr/csm-1b-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 csm 1b use?
Hugging Face lists the license as apache-2.0. 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 CSM-1B on 8 GB VRAM?
Yes, the Q4_K quantized file is 1.44 GB and requires about 2.1 GB of VRAM at 8 k context.
Does CSM-1B support multiple languages?
The model card does not list supported languages; use the base model’s language capabilities.
Is CSM-1B licensed for commercial use?
It is released under Apache‑2.0, which permits commercial use with attribution.