| | --- |
| | license: cc-by-nc-4.0 |
| | tags: |
| | - merge |
| | - mergekit |
| | - lazymergekit |
| | base_model: |
| | - liminerity/M7-7b |
| | - rwitz/experiment26-truthy-iter-0 |
| | quantized_by: bartowski |
| | pipeline_tag: text-generation |
| | --- |
| | |
| | ## Llamacpp Quantizations of Zebrafish-7B |
| |
|
| | Using <a href="https://github.com/ggerganov/llama.cpp/">llama.cpp</a> release <a href="https://github.com/ggerganov/llama.cpp/releases/tag/b2536">b2536</a> for quantization. |
| |
|
| | Original model: https://huggingface.co/mlabonne/Zebrafish-7B |
| |
|
| | Download a file (not the whole branch) from below: |
| |
|
| | | Filename | Quant type | File Size | Description | |
| | | -------- | ---------- | --------- | ----------- | |
| | | [Zebrafish-7B-Q8_0.gguf](https://huggingface.co/bartowski/Zebrafish-7B-GGUF/blob/main/Zebrafish-7B-Q8_0.gguf) | Q8_0 | 7.69GB | Extremely high quality, generally unneeded but max available quant. | |
| | | [Zebrafish-7B-Q6_K.gguf](https://huggingface.co/bartowski/Zebrafish-7B-GGUF/blob/main/Zebrafish-7B-Q6_K.gguf) | Q6_K | 5.94GB | Very high quality, near perfect, *recommended*. | |
| | | [Zebrafish-7B-Q5_K_M.gguf](https://huggingface.co/bartowski/Zebrafish-7B-GGUF/blob/main/Zebrafish-7B-Q5_K_M.gguf) | Q5_K_M | 5.13GB | High quality, very usable. | |
| | | [Zebrafish-7B-Q5_K_S.gguf](https://huggingface.co/bartowski/Zebrafish-7B-GGUF/blob/main/Zebrafish-7B-Q5_K_S.gguf) | Q5_K_S | 4.99GB | High quality, very usable. | |
| | | [Zebrafish-7B-Q5_0.gguf](https://huggingface.co/bartowski/Zebrafish-7B-GGUF/blob/main/Zebrafish-7B-Q5_0.gguf) | Q5_0 | 4.99GB | High quality, older format, generally not recommended. | |
| | | [Zebrafish-7B-Q4_K_M.gguf](https://huggingface.co/bartowski/Zebrafish-7B-GGUF/blob/main/Zebrafish-7B-Q4_K_M.gguf) | Q4_K_M | 4.36GB | Good quality, uses about 4.83 bits per weight. | |
| | | [Zebrafish-7B-Q4_K_S.gguf](https://huggingface.co/bartowski/Zebrafish-7B-GGUF/blob/main/Zebrafish-7B-Q4_K_S.gguf) | Q4_K_S | 4.14GB | Slightly lower quality with small space savings. | |
| | | [Zebrafish-7B-IQ4_NL.gguf](https://huggingface.co/bartowski/Zebrafish-7B-GGUF/blob/main/Zebrafish-7B-IQ4_NL.gguf) | IQ4_NL | 4.15GB | Decent quality, similar to Q4_K_S, new method of quanting, | |
| | | [Zebrafish-7B-IQ4_XS.gguf](https://huggingface.co/bartowski/Zebrafish-7B-GGUF/blob/main/Zebrafish-7B-IQ4_XS.gguf) | IQ4_XS | 3.94GB | Decent quality, new method with similar performance to Q4. | |
| | | [Zebrafish-7B-Q4_0.gguf](https://huggingface.co/bartowski/Zebrafish-7B-GGUF/blob/main/Zebrafish-7B-Q4_0.gguf) | Q4_0 | 4.10GB | Decent quality, older format, generally not recommended. | |
| | | [Zebrafish-7B-Q3_K_L.gguf](https://huggingface.co/bartowski/Zebrafish-7B-GGUF/blob/main/Zebrafish-7B-Q3_K_L.gguf) | Q3_K_L | 3.82GB | Lower quality but usable, good for low RAM availability. | |
| | | [Zebrafish-7B-Q3_K_M.gguf](https://huggingface.co/bartowski/Zebrafish-7B-GGUF/blob/main/Zebrafish-7B-Q3_K_M.gguf) | Q3_K_M | 3.51GB | Even lower quality. | |
| | | [Zebrafish-7B-IQ3_M.gguf](https://huggingface.co/bartowski/Zebrafish-7B-GGUF/blob/main/Zebrafish-7B-IQ3_M.gguf) | IQ3_M | 3.28GB | Medium-low quality, new method with decent performance. | |
| | | [Zebrafish-7B-IQ3_S.gguf](https://huggingface.co/bartowski/Zebrafish-7B-GGUF/blob/main/Zebrafish-7B-IQ3_S.gguf) | IQ3_S | 3.18GB | Lower quality, new method with decent performance, recommended over Q3 quants. | |
| | | [Zebrafish-7B-Q3_K_S.gguf](https://huggingface.co/bartowski/Zebrafish-7B-GGUF/blob/main/Zebrafish-7B-Q3_K_S.gguf) | Q3_K_S | 3.16GB | Low quality, not recommended. | |
| | | [Zebrafish-7B-Q2_K.gguf](https://huggingface.co/bartowski/Zebrafish-7B-GGUF/blob/main/Zebrafish-7B-Q2_K.gguf) | Q2_K | 2.71GB | Extremely low quality, *not* recommended. |
| | |
| | Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski |
| | |
