GigaAM v3 E2E-RNN-T needs about 0.7 GB of memory at Q4_K_M (a 0.2 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 6 quants, from 0.2 GB (Q4_K_M) to 0.9 GB (F32).
GigaAM v3 E2E-RNN-T: 223 M‑parameter Russian ASR model
GigaAM v3 E2E‑RNN‑T is a 223 million‑parameter Conformer encoder with an RNN‑T transducer head. It outputs cased Russian text with punctuation and normalization. The model is trained on diverse Russian speech and performs well on new domains.
Good at
- End‑to‑end Russian speech‑to‑text
- Punctuation and text normalization
- Offline decoding with greedy RNN‑T
Limits
- No vision or translation support
- Requires 8 k context for memory estimates
- MIT license only
Best for
- Transcribing Russian podcasts
- Generating subtitles for Russian videos
- Batch processing of Russian audio files
Who it’s for. Developers building offline Russian speech‑to‑text pipelines. Researchers needing a lightweight, end‑to‑end RNN‑T model for Russian audio. Audio‑processing teams that prefer non‑streaming, greedy decoding.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K_M | 0.2 GB | 0.7 GB | Fits | Get file → | |
| Q5_K_M | 0.2 GB | 0.7 GB | Fits | Get file → | |
| Q6_K | 0.2 GB | 0.8 GB | Fits | Get file → | |
| Q8_0 | 0.3 GB | 0.8 GB | Fits | Get file → | |
| F16 | 0.5 GB | 1.0 GB | Fits | Get file → | |
| F32 | 0.9 GB | 1.5 GB | Fits | Get file → |
For 8 GB VRAM, use Q4_K_M or Q5_K_M (0.18 GB or 0.21 GB). For 4 GB VRAM, Q4_K_M is recommended. If more memory is available, Q6_K or Q8_0 provide higher precision. Full‑precision F16 and F32 require 1–1.5 GB VRAM at 8 k context.
What GPU do you need for GigaAM v3 E2E-RNN-T?
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 F32 (0.9 GB file, about 1.5 GB in use).
Which GigaAM v3 E2E-RNN-T quant should you choose?
Q4_K_M is the usual starting point for GigaAM v3 E2E-RNN-T: it balances file size and quality well. Here it is a 0.2 GB file.
The Q8_0 file is 0.3 GB, about 1.5 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.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about GigaAM v3 E2E-RNN-T
How much VRAM does GigaAM v3 E2E-RNN-T need?
About 0.7 GB at Q4_K_M with an 8K context window. That is the 0.2 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 GigaAM v3 E2E-RNN-T on an 8 GB GPU?
Yes, with F32: the 0.9 GB file needs about 1.5 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of GigaAM v3 E2E-RNN-T GGUF?
It depends on the quant: Q4_K_M is 0.2 GB, Q5_K_M is 0.2 GB, Q6_K is 0.2 GB, Q8_0 is 0.3 GB, F16 is 0.5 GB, F32 is 0.9 GB.
Where can I download GigaAM v3 E2E-RNN-T GGUF?
The files are on Hugging Face in the handy-computer/gigaam-v3-e2e-rnnt-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 GigaAM v3 E2E-RNN-T use?
Hugging Face lists the license as mit. 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 GigaAM v3 E2E‑RNN‑T on 8 GB VRAM?
Yes, the Q4_K_M or Q5_K_M quantizations fit within 8 GB VRAM with an estimated 0.7 GB usage at 8 k context.
Does the model support streaming inference?
No, the model is not streaming and processes audio in overlapping windows for inputs longer than 25 s.
What license applies to this model?
The model is released under the MIT license.