mradermacher/BanglaLLama-3-8b-Bangla2B-base-v0.0.1-GGUF

1---2base_model: BanglaLLM/BanglaLLama-3-8b-Bangla2B-base-v0.0.13language:4- bn5- en6library_name: transformers7license: llama38mradermacher:9  readme_rev: 110quantized_by: mradermacher11tags:12- bengali13- bangalnlp14- nlp15- bangla16- bangladesh17- transformer18- LLama19- banglallama20---21## About22 23<!-- ### quantize_version: 2 -->24<!-- ### output_tensor_quantised: 1 -->25<!-- ### convert_type: hf -->26<!-- ### vocab_type:  -->27<!-- ### tags:  -->28static quants of https://huggingface.co/BanglaLLM/BanglaLLama-3-8b-Bangla2B-base-v0.0.129 30<!-- provided-files -->31 32***For a convenient overview and download list, visit our [model page for this model](https://hf.tst.eu/model#BanglaLLama-3-8b-Bangla2B-base-v0.0.1-GGUF).***33 34weighted/imatrix quants seem not to be available (by me) at this time. If they do not show up a week or so after the static ones, I have probably not planned for them. Feel free to request them by opening a Community Discussion.35## Usage36 37If you are unsure how to use GGUF files, refer to one of [TheBloke's38READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for39more details, including on how to concatenate multi-part files.40 41## Provided Quants42 43(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)44 45| Link | Type | Size/GB | Notes |46|:-----|:-----|--------:|:------|47| [GGUF](https://huggingface.co/mradermacher/BanglaLLama-3-8b-Bangla2B-base-v0.0.1-GGUF/resolve/main/BanglaLLama-3-8b-Bangla2B-base-v0.0.1.Q2_K.gguf) | Q2_K | 3.3 |  |48| [GGUF](https://huggingface.co/mradermacher/BanglaLLama-3-8b-Bangla2B-base-v0.0.1-GGUF/resolve/main/BanglaLLama-3-8b-Bangla2B-base-v0.0.1.Q3_K_S.gguf) | Q3_K_S | 3.8 |  |49| [GGUF](https://huggingface.co/mradermacher/BanglaLLama-3-8b-Bangla2B-base-v0.0.1-GGUF/resolve/main/BanglaLLama-3-8b-Bangla2B-base-v0.0.1.Q3_K_M.gguf) | Q3_K_M | 4.1 | lower quality |50| [GGUF](https://huggingface.co/mradermacher/BanglaLLama-3-8b-Bangla2B-base-v0.0.1-GGUF/resolve/main/BanglaLLama-3-8b-Bangla2B-base-v0.0.1.Q3_K_L.gguf) | Q3_K_L | 4.4 |  |51| [GGUF](https://huggingface.co/mradermacher/BanglaLLama-3-8b-Bangla2B-base-v0.0.1-GGUF/resolve/main/BanglaLLama-3-8b-Bangla2B-base-v0.0.1.IQ4_XS.gguf) | IQ4_XS | 4.6 |  |52| [GGUF](https://huggingface.co/mradermacher/BanglaLLama-3-8b-Bangla2B-base-v0.0.1-GGUF/resolve/main/BanglaLLama-3-8b-Bangla2B-base-v0.0.1.Q4_K_S.gguf) | Q4_K_S | 4.8 | fast, recommended |53| [GGUF](https://huggingface.co/mradermacher/BanglaLLama-3-8b-Bangla2B-base-v0.0.1-GGUF/resolve/main/BanglaLLama-3-8b-Bangla2B-base-v0.0.1.Q4_K_M.gguf) | Q4_K_M | 5.0 | fast, recommended |54| [GGUF](https://huggingface.co/mradermacher/BanglaLLama-3-8b-Bangla2B-base-v0.0.1-GGUF/resolve/main/BanglaLLama-3-8b-Bangla2B-base-v0.0.1.Q5_K_S.gguf) | Q5_K_S | 5.7 |  |55| [GGUF](https://huggingface.co/mradermacher/BanglaLLama-3-8b-Bangla2B-base-v0.0.1-GGUF/resolve/main/BanglaLLama-3-8b-Bangla2B-base-v0.0.1.Q5_K_M.gguf) | Q5_K_M | 5.8 |  |56| [GGUF](https://huggingface.co/mradermacher/BanglaLLama-3-8b-Bangla2B-base-v0.0.1-GGUF/resolve/main/BanglaLLama-3-8b-Bangla2B-base-v0.0.1.Q6_K.gguf) | Q6_K | 6.7 | very good quality |57| [GGUF](https://huggingface.co/mradermacher/BanglaLLama-3-8b-Bangla2B-base-v0.0.1-GGUF/resolve/main/BanglaLLama-3-8b-Bangla2B-base-v0.0.1.Q8_0.gguf) | Q8_0 | 8.6 | fast, best quality |58| [GGUF](https://huggingface.co/mradermacher/BanglaLLama-3-8b-Bangla2B-base-v0.0.1-GGUF/resolve/main/BanglaLLama-3-8b-Bangla2B-base-v0.0.1.f16.gguf) | f16 | 16.2 | 16 bpw, overkill |59 60Here is a handy graph by ikawrakow comparing some lower-quality quant61types (lower is better):62 63![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)64 65And here are Artefact2's thoughts on the matter:66https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec967 68## FAQ / Model Request69 70See https://huggingface.co/mradermacher/model_requests for some answers to71questions you might have and/or if you want some other model quantized.72 73## Thanks74 75I thank my company, [nethype GmbH](https://www.nethype.de/), for letting76me use its servers and providing upgrades to my workstation to enable77this work in my free time.78 79<!-- end -->80