Add quantized mixtral support (#2673)
This commit is contained in:
@@ -30,6 +30,7 @@ _MODELS = {
|
||||
"LLaMAForCausalLM": ("llama", "LlamaForCausalLM"),
|
||||
"MistralForCausalLM": ("mistral", "MistralForCausalLM"),
|
||||
"MixtralForCausalLM": ("mixtral", "MixtralForCausalLM"),
|
||||
"QuantMixtralForCausalLM": ("mixtral_quant", "MixtralForCausalLM"),
|
||||
# transformers's mpt class has lower case
|
||||
"MptForCausalLM": ("mpt", "MPTForCausalLM"),
|
||||
"MPTForCausalLM": ("mpt", "MPTForCausalLM"),
|
||||
|
||||
Reference in New Issue
Block a user