pe-nlp
/

deepseek-pt-v3-quantized-int4

Text Generation

Inference Endpoints

4-bit precision

Model card Files Files and versions Community

deepseek-pt-v3-quantized-int4 / generation_config.json

shawnzzzzz's picture

AutoGPTQ model for pe-nlp/deepseek-pt-v3-quantized-int4: 8bits, gr128, desc_act=False

d1641d3 verified 6 months ago

history blame contribute delete

119 Bytes

	{
	"_from_model_config": true,
	"bos_token_id": 32013,
	"eos_token_id": 32021,
	"transformers_version": "4.38.2"
	}