---
language:
- en
library_name: transformers
pipeline_tag: text-generation
datasets:
- teknium/OpenHermes-2.5
- TokenBender/python_eval_instruct_51k
- codefuse-ai/Evol-instruction-66k
tags:
- code
license: apache-2.0
model-index:
- name: SpeechlessCoder
  results:
  - task:
      type: text-generation
    dataset:
      type: openai_humaneval
      name: HumanEval
    metrics:
    - name: pass@1
      type: pass@1
      value: 0.0
      verified: false
quantized_by: bartowski
---

## Llamacpp Quantizations of speechless-starcoder2-7b

Using <a href="https://github.com/ggerganov/llama.cpp/">llama.cpp</a> release <a href="https://github.com/ggerganov/llama.cpp/releases/tag/b2354">b2354</a> for quantization.

Original model: https://huggingface.co/uukuguy/speechless-starcoder2-7b

Download a file (not the whole branch) from below:

| Filename | Quant type | File Size | Description |
| -------- | ---------- | --------- | ----------- |
| [speechless-starcoder2-7b-Q8_0.gguf](https://huggingface.co/bartowski/speechless-starcoder2-7b-GGUF//main/speechless-starcoder2-7b-Q8_0.gguf) | Q8_0 | 7.62GB | Extremely high quality, generally unneeded but max available quant. |
| [speechless-starcoder2-7b-Q6_K.gguf](https://huggingface.co/bartowski/speechless-starcoder2-7b-GGUF//main/speechless-starcoder2-7b-Q6_K.gguf) | Q6_K | 5.89GB | Very high quality, near perfect, *recommended*. |
| [speechless-starcoder2-7b-Q5_K_M.gguf](https://huggingface.co/bartowski/speechless-starcoder2-7b-GGUF//main/speechless-starcoder2-7b-Q5_K_M.gguf) | Q5_K_M | 5.12GB | High quality, very usable. |
| [speechless-starcoder2-7b-Q5_K_S.gguf](https://huggingface.co/bartowski/speechless-starcoder2-7b-GGUF//main/speechless-starcoder2-7b-Q5_K_S.gguf) | Q5_K_S | 4.93GB | High quality, very usable. |
| [speechless-starcoder2-7b-Q5_0.gguf](https://huggingface.co/bartowski/speechless-starcoder2-7b-GGUF//main/speechless-starcoder2-7b-Q5_0.gguf) | Q5_0 | 4.93GB | High quality, older format, generally not recommended. |
| [speechless-starcoder2-7b-Q4_K_M.gguf](https://huggingface.co/bartowski/speechless-starcoder2-7b-GGUF//main/speechless-starcoder2-7b-Q4_K_M.gguf) | Q4_K_M | 4.40GB | Good quality, similar to 4.25 bpw. |
| [speechless-starcoder2-7b-Q4_K_S.gguf](https://huggingface.co/bartowski/speechless-starcoder2-7b-GGUF//main/speechless-starcoder2-7b-Q4_K_S.gguf) | Q4_K_S | 4.12GB | Slightly lower quality with small space savings. |
| [speechless-starcoder2-7b-Q4_0.gguf](https://huggingface.co/bartowski/speechless-starcoder2-7b-GGUF//main/speechless-starcoder2-7b-Q4_0.gguf) | Q4_0 | 4.04GB | Decent quality, older format, generally not recommended. |
| [speechless-starcoder2-7b-Q3_K_L.gguf](https://huggingface.co/bartowski/speechless-starcoder2-7b-GGUF//main/speechless-starcoder2-7b-Q3_K_L.gguf) | Q3_K_L | 3.98GB | Lower quality but usable, good for low RAM availability. |
| [speechless-starcoder2-7b-Q3_K_M.gguf](https://huggingface.co/bartowski/speechless-starcoder2-7b-GGUF//main/speechless-starcoder2-7b-Q3_K_M.gguf) | Q3_K_M | 3.59GB | Even lower quality. |
| [speechless-starcoder2-7b-Q3_K_S.gguf](https://huggingface.co/bartowski/speechless-starcoder2-7b-GGUF//main/speechless-starcoder2-7b-Q3_K_S.gguf) | Q3_K_S | 3.09GB | Low quality, not recommended. |
| [speechless-starcoder2-7b-Q2_K.gguf](https://huggingface.co/bartowski/speechless-starcoder2-7b-GGUF//main/speechless-starcoder2-7b-Q2_K.gguf) | Q2_K | 2.72GB | Extremely low quality, *not* recommended.

Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski