Views
No views yet
| Name | Quant method | Size |
|---|---|---|
| Mixtral-GQA-400m-v2.Q2_K.gguf | Q2_K | 0.72GB |
| Mixtral-GQA-400m-v2.IQ3_XS.gguf | IQ3_XS | 0.8GB |
| Mixtral-GQA-400m-v2.IQ3_S.gguf | IQ3_S | 0.84GB |
| Mixtral-GQA-400m-v2.Q3_K_S.gguf | Q3_K_S | 0.84GB |
| Mixtral-GQA-400m-v2.IQ3_M.gguf | IQ3_M | 0.87GB |
| Mixtral-GQA-400m-v2.Q3_K.gguf | Q3_K | 0.92GB |
| Mixtral-GQA-400m-v2.Q3_K_M.gguf | Q3_K_M | 0.92GB |
| Mixtral-GQA-400m-v2.Q3_K_L.gguf | Q3_K_L | 0.98GB |
| Mixtral-GQA-400m-v2.IQ4_XS.gguf | IQ4_XS | 1.02GB |
| Mixtral-GQA-400m-v2.Q4_0.gguf | Q4_0 | 1.07GB |
| Mixtral-GQA-400m-v2.IQ4_NL.gguf | IQ4_NL | 1.08GB |
| Mixtral-GQA-400m-v2.Q4_K_S.gguf | Q4_K_S | 1.08GB |
| Mixtral-GQA-400m-v2.Q4_K.gguf | Q4_K | 1.12GB |
| Mixtral-GQA-400m-v2.Q4_K_M.gguf | Q4_K_M | 1.12GB |
| Mixtral-GQA-400m-v2.Q4_1.gguf | Q4_1 | 1.19GB |
| Mixtral-GQA-400m-v2.Q5_0.gguf | Q5_0 | 1.3GB |
| Mixtral-GQA-400m-v2.Q5_K_S.gguf | Q5_K_S | 1.3GB |
| Mixtral-GQA-400m-v2.Q5_K.gguf | Q5_K | 1.32GB |
| Mixtral-GQA-400m-v2.Q5_K_M.gguf | Q5_K_M | 1.32GB |
| Mixtral-GQA-400m-v2.Q5_1.gguf | Q5_1 | 1.41GB |
| Mixtral-GQA-400m-v2.Q6_K.gguf | Q6_K | 1.54GB |
| Mixtral-GQA-400m-v2.Q8_0.gguf | Q8_0 | 1.99GB |
1# !pip install -U -q transformers datasets accelerate sentencepiece
2import pprint as pp
3from transformers import pipeline
4
5pipe = pipeline(
6 "text-generation",
7 model="BEE-spoke-data/Mixtral-GQA-400m-v2",
8 device_map="auto",
9)
10pipe.model.config.pad_token_id = pipe.model.config.eos_token_id
11
12prompt = "My favorite movie is Godfather because"
13
14res = pipe(
15 prompt,
16 max_new_tokens=256,
17 top_k=4,
18 penalty_alpha=0.6,
19 use_cache=True,
20 no_repeat_ngram_size=4,
21 repetition_penalty=1.1,
22 renormalize_logits=True,
23)
24pp.pprint(res[0])