|
|
@ -127,7 +127,7 @@ llama2_70b = Model(
|
|
|
|
codellama_34b_instruct = Model(
|
|
|
|
codellama_34b_instruct = Model(
|
|
|
|
name = "codellama/CodeLlama-34b-Instruct-hf",
|
|
|
|
name = "codellama/CodeLlama-34b-Instruct-hf",
|
|
|
|
base_provider = "meta",
|
|
|
|
base_provider = "meta",
|
|
|
|
best_provider = RetryProvider([HuggingChat, DeepInfra])
|
|
|
|
best_provider = HuggingChat
|
|
|
|
)
|
|
|
|
)
|
|
|
|
|
|
|
|
|
|
|
|
codellama_70b_instruct = Model(
|
|
|
|
codellama_70b_instruct = Model(
|
|
|
@ -146,9 +146,21 @@ mixtral_8x7b = Model(
|
|
|
|
mistral_7b = Model(
|
|
|
|
mistral_7b = Model(
|
|
|
|
name = "mistralai/Mistral-7B-Instruct-v0.1",
|
|
|
|
name = "mistralai/Mistral-7B-Instruct-v0.1",
|
|
|
|
base_provider = "huggingface",
|
|
|
|
base_provider = "huggingface",
|
|
|
|
best_provider = RetryProvider([DeepInfra, HuggingChat, HuggingFace, PerplexityLabs])
|
|
|
|
best_provider = RetryProvider([HuggingChat, HuggingFace, PerplexityLabs])
|
|
|
|
|
|
|
|
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
mistral_7b_v02 = Model(
|
|
|
|
|
|
|
|
name = "mistralai/Mistral-7B-Instruct-v0.2",
|
|
|
|
|
|
|
|
base_provider = "huggingface",
|
|
|
|
|
|
|
|
best_provider = DeepInfra
|
|
|
|
)
|
|
|
|
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
# mixtral_8x22b = Model(
|
|
|
|
|
|
|
|
# name = "mistralai/Mixtral-8x22B-v0.1",
|
|
|
|
|
|
|
|
# base_provider = "huggingface",
|
|
|
|
|
|
|
|
# best_provider = DeepInfra
|
|
|
|
|
|
|
|
# )
|
|
|
|
|
|
|
|
|
|
|
|
# Misc models
|
|
|
|
# Misc models
|
|
|
|
dolphin_mixtral_8x7b = Model(
|
|
|
|
dolphin_mixtral_8x7b = Model(
|
|
|
|
name = "cognitivecomputations/dolphin-2.6-mixtral-8x7b",
|
|
|
|
name = "cognitivecomputations/dolphin-2.6-mixtral-8x7b",
|
|
|
@ -168,12 +180,6 @@ airoboros_70b = Model(
|
|
|
|
best_provider = DeepInfra
|
|
|
|
best_provider = DeepInfra
|
|
|
|
)
|
|
|
|
)
|
|
|
|
|
|
|
|
|
|
|
|
airoboros_l2_70b = Model(
|
|
|
|
|
|
|
|
name = "jondurbin/airoboros-l2-70b-gpt4-1.4.1",
|
|
|
|
|
|
|
|
base_provider = "huggingface",
|
|
|
|
|
|
|
|
best_provider = DeepInfra
|
|
|
|
|
|
|
|
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
openchat_35 = Model(
|
|
|
|
openchat_35 = Model(
|
|
|
|
name = "openchat/openchat_3.5",
|
|
|
|
name = "openchat/openchat_3.5",
|
|
|
|
base_provider = "huggingface",
|
|
|
|
base_provider = "huggingface",
|
|
|
@ -290,10 +296,11 @@ class ModelUtils:
|
|
|
|
|
|
|
|
|
|
|
|
'mixtral-8x7b': mixtral_8x7b,
|
|
|
|
'mixtral-8x7b': mixtral_8x7b,
|
|
|
|
'mistral-7b': mistral_7b,
|
|
|
|
'mistral-7b': mistral_7b,
|
|
|
|
|
|
|
|
'mistral-7b-v02': mistral_7b_v02,
|
|
|
|
|
|
|
|
# 'mixtral-8x22b': mixtral_8x22b,
|
|
|
|
'dolphin-mixtral-8x7b': dolphin_mixtral_8x7b,
|
|
|
|
'dolphin-mixtral-8x7b': dolphin_mixtral_8x7b,
|
|
|
|
'lzlv-70b': lzlv_70b,
|
|
|
|
'lzlv-70b': lzlv_70b,
|
|
|
|
'airoboros-70b': airoboros_70b,
|
|
|
|
'airoboros-70b': airoboros_70b,
|
|
|
|
'airoboros-l2-70b': airoboros_l2_70b,
|
|
|
|
|
|
|
|
'openchat_3.5': openchat_35,
|
|
|
|
'openchat_3.5': openchat_35,
|
|
|
|
'gemini': gemini,
|
|
|
|
'gemini': gemini,
|
|
|
|
'gemini-pro': gemini_pro,
|
|
|
|
'gemini-pro': gemini_pro,
|
|
|
|