Gemma 2 Mitra E needs about 9.1 GB of memory at Q4_K_M (a 5.8 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 12 quants, from 3.8 GB (Q2_K) to 18.5 GB (F16).
Gemma 2 Mitra E – 9.24B multilingual sentence embedding
Gemma 2 Mitra E is a 9.24B parameter encoder designed for multilingual sentence embeddings. It produces L2‑normalized vectors for semantic similarity and retrieval across languages such as Sanskrit, Tibetan, Pali, Chinese, and English. The model is optimized for embedding tasks and does not support vision or instruction generation.
Good at
- Multilingual sentence embeddings
- Semantic similarity search
- Cross‑lingual retrieval
Limits
- No vision support
- No instruction generation
- License not specified
Best for
- Embedding queries for RAG
- Aligning Buddhist texts across languages
- Nearest‑neighbor search in multilingual corpora
Who it’s for. Developers building multilingual search or alignment pipelines that require high‑quality sentence vectors. Researchers needing a lightweight embedding model for cross‑lingual similarity tasks.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 3.8 GB | 7.1 GB | Fits | Get file → | |
| Q3_K_S | 4.3 GB | 7.7 GB | Tight | Get file → | |
| Q3_K_M | 4.8 GB | 8.1 GB | Too big | Get file → | |
| Q3_K_L | 5.1 GB | 8.5 GB | Too big | Get file → | |
| IQ4_XS | 5.2 GB | 8.5 GB | Too big | Get file → | |
| Q4_K_S | 5.5 GB | 8.8 GB | Too big | Get file → | |
| Q4_K_M | 5.8 GB | 9.1 GB | Too big | Get file → | |
| Q5_K_S | 6.5 GB | 9.8 GB | Too big | Get file → | |
| Q5_K_M | 6.6 GB | 10.0 GB | Too big | Get file → | |
| Q6_K | 7.6 GB | 10.9 GB | Too big | Get file → | |
| Q8_0 | 9.8 GB | 13.1 GB | Too big | Get file → | |
| F16 | 18.5 GB | 21.8 GB | Too big | Get file → |
For 8 GB VRAM, use Q4_K_S (5.48 GB, 8.8 GB memory) or Q4_K_M (5.76 GB, 9.1 GB memory). If 12 GB VRAM is available, Q5_K_S (6.48 GB, 9.8 GB memory) or Q5_K_M (6.65 GB, 10 GB memory) offer better quality. Larger VRAM allows Q6_K or Q8_0 for higher precision.
What GPU do you need for Gemma 2 Mitra E?
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 Q3_K_S (4.3 GB file, about 7.7 GB in use).
- 12 GB VRAM: the best quant that fits is Q6_K (7.6 GB file, about 10.9 GB in use).
- 16 GB VRAM: the best quant that fits is Q8_0 (9.8 GB file, about 13.1 GB in use).
- 24 GB VRAM: the best quant that fits is F16 (18.5 GB file, about 21.8 GB in use).
Which Gemma 2 Mitra E quant should you choose?
Q4_K_M is the usual starting point for Gemma 2 Mitra E: it balances file size and quality well. Here it is a 5.8 GB file.
The Q8_0 file is 9.8 GB, about 1.7 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 Q2_K at 3.8 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
Source
Frequently asked questions about Gemma 2 Mitra E
How much VRAM does Gemma 2 Mitra E need?
About 9.1 GB at Q4_K_M with an 8K context window. That is the 5.8 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 Gemma 2 Mitra E on an 8 GB GPU?
Yes, with Q3_K_S: the 4.3 GB file needs about 7.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Gemma 2 Mitra E GGUF?
It depends on the quant: Q2_K is 3.8 GB, Q3_K_S is 4.3 GB, Q3_K_M is 4.8 GB, Q3_K_L is 5.1 GB, IQ4_XS is 5.2 GB, Q4_K_S is 5.5 GB and 6 more.
What context length does Gemma 2 Mitra E support?
The GGUF header lists a trained context length of 8,192 tokens (8K). The KV cache grows with the context you use: about 2.8 GB at 8K and 2.8 GB at 8K.
Where can I download Gemma 2 Mitra E GGUF?
The files are on Hugging Face in the mradermacher/gemma-2-mitra-e-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
Can I run Gemma 2 Mitra E on 8 GB VRAM?
Yes, using Q4_K_S or Q4_K_M quants will fit within 8 GB VRAM, with memory usage around 8.8–9.1 GB at 8k context.
Does this model support image inputs?
No, Gemma 2 Mitra E is an encoder for text only and does not handle vision tasks.
What prompt format should I use for queries?
Use the asymmetric format: "Please find the semantically most similar text in {language}.\n {sentence_text}" for queries and raw text for corpus.