You can deploy this model and start earning money today!

RichardErkhov/
nnethercott_-_llava-v1.5-7b-hf-vicuna-gguf

Name: nnethercott_-_llava-v1.5-7b-hf-vicuna-gguf
Availability: InStock

endpoints_compatible

gguf

template

Views

No views yet

Quantization made by Richard Erkhov.

Github

Discord

Request more models

llava-v1.5-7b-hf-vicuna - GGUF

Model creator: https://huggingface.co/nnethercott/
Original model: https://huggingface.co/nnethercott/llava-v1.5-7b-hf-vicuna/

Name	Quant method	Size
llava-v1.5-7b-hf-vicuna.Q2_K.gguf	Q2_K	2.36GB
llava-v1.5-7b-hf-vicuna.IQ3_XS.gguf	IQ3_XS	2.6GB
llava-v1.5-7b-hf-vicuna.IQ3_S.gguf	IQ3_S	2.75GB
llava-v1.5-7b-hf-vicuna.Q3_K_S.gguf	Q3_K_S	2.75GB
llava-v1.5-7b-hf-vicuna.IQ3_M.gguf	IQ3_M	2.9GB
llava-v1.5-7b-hf-vicuna.Q3_K.gguf	Q3_K	3.07GB
llava-v1.5-7b-hf-vicuna.Q3_K_M.gguf	Q3_K_M	3.07GB
llava-v1.5-7b-hf-vicuna.Q3_K_L.gguf	Q3_K_L	3.35GB
llava-v1.5-7b-hf-vicuna.IQ4_XS.gguf	IQ4_XS	3.4GB
llava-v1.5-7b-hf-vicuna.Q4_0.gguf	Q4_0	3.56GB
llava-v1.5-7b-hf-vicuna.IQ4_NL.gguf	IQ4_NL	3.58GB
llava-v1.5-7b-hf-vicuna.Q4_K_S.gguf	Q4_K_S	3.59GB
llava-v1.5-7b-hf-vicuna.Q4_K.gguf	Q4_K	3.8GB
llava-v1.5-7b-hf-vicuna.Q4_K_M.gguf	Q4_K_M	3.8GB
llava-v1.5-7b-hf-vicuna.Q4_1.gguf	Q4_1	3.95GB
llava-v1.5-7b-hf-vicuna.Q5_0.gguf	Q5_0	4.33GB
llava-v1.5-7b-hf-vicuna.Q5_K_S.gguf	Q5_K_S	4.33GB
llava-v1.5-7b-hf-vicuna.Q5_K.gguf	Q5_K	4.46GB
llava-v1.5-7b-hf-vicuna.Q5_K_M.gguf	Q5_K_M	4.46GB
llava-v1.5-7b-hf-vicuna.Q5_1.gguf	Q5_1	4.72GB
llava-v1.5-7b-hf-vicuna.Q6_K.gguf	Q6_K	5.15GB
llava-v1.5-7b-hf-vicuna.Q8_0.gguf	Q8_0	6.67GB

Original model description:

license: llama2 model-index:

name: llava-v1.5-7b-hf-vicuna results:
- task: type: text-generation name: Text Generation dataset: name: AI2 Reasoning Challenge (25-Shot) type: ai2_arc config: ARC-Challenge split: test args: num_few_shot: 25 metrics:
  - type: acc_norm value: 52.65 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=nnethercott/llava-v1.5-7b-hf-vicuna name: Open LLM Leaderboard
- task: type: text-generation name: Text Generation dataset: name: HellaSwag (10-Shot) type: hellaswag split: validation args: num_few_shot: 10 metrics:
  - type: acc_norm value: 76.09 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=nnethercott/llava-v1.5-7b-hf-vicuna name: Open LLM Leaderboard
- task: type: text-generation name: Text Generation dataset: name: MMLU (5-Shot) type: cais/mmlu config: all split: test args: num_few_shot: 5 metrics:
  - type: acc value: 51.68 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=nnethercott/llava-v1.5-7b-hf-vicuna name: Open LLM Leaderboard
- task: type: text-generation name: Text Generation dataset: name: TruthfulQA (0-shot) type: truthful_qa config: multiple_choice split: validation args: num_few_shot: 0 metrics:
  - type: mc2 value: 45.86 source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=nnethercott/llava-v1.5-7b-hf-vicuna name: Open LLM Leaderboard
- task: type: text-generation name: Text Generation dataset: name: Winogrande (5-shot) type: winogrande config: winogrande_xl split: validation args: num_few_shot: 5 metrics:
  - type: acc value: 72.06 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=nnethercott/llava-v1.5-7b-hf-vicuna name: Open LLM Leaderboard
- task: type: text-generation name: Text Generation dataset: name: GSM8k (5-shot) type: gsm8k config: main split: test args: num_few_shot: 5 metrics:
  - type: acc value: 15.31 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=nnethercott/llava-v1.5-7b-hf-vicuna name: Open LLM Leaderboard

Model details

Motivation This models contains the fine-tuned weights from llava-hf/llava-1.5-7b-hf so LLM benchmarking can be done.

Model type: LLaVA is an open-source chatbot trained by fine-tuning LLaMA/Vicuna on GPT-generated multimodal instruction-following data. It is an auto-regressive language model, based on the transformer architecture.

License

Training dataset

558K filtered image-text pairs from LAION/CC/SBU, captioned by BLIP.
158K GPT-generated multimodal instruction-following data.
450K academic-task-oriented VQA data mixture.
40K ShareGPT data.

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	52.28
AI2 Reasoning Challenge (25-Shot)	52.65
HellaSwag (10-Shot)	76.09
MMLU (5-Shot)	51.68
TruthfulQA (0-shot)	45.86
Winogrande (5-shot)	72.06
GSM8k (5-shot)	15.31

RichardErkhov/nnethercott_-_llava-v1.5-7b-hf-vicuna-gguf