mirror of
https://github.com/xtekky/gpt4free.git
synced 2024-11-27 05:25:33 +03:00
315 lines
7.5 KiB
Python
315 lines
7.5 KiB
Python
from __future__ import annotations
|
|
|
|
from dataclasses import dataclass
|
|
|
|
from .Provider import RetryProvider, ProviderType
|
|
from .Provider import (
|
|
Chatgpt4Online,
|
|
PerplexityLabs,
|
|
GeminiProChat,
|
|
ChatgptNext,
|
|
HuggingChat,
|
|
HuggingFace,
|
|
ChatgptDemo,
|
|
GptForLove,
|
|
ChatgptAi,
|
|
DeepInfra,
|
|
ChatBase,
|
|
GigaChat,
|
|
Liaobots,
|
|
FreeGpt,
|
|
Llama2,
|
|
Vercel,
|
|
Gemini,
|
|
GptGo,
|
|
Gpt6,
|
|
Bing,
|
|
You,
|
|
Pi,
|
|
)
|
|
|
|
@dataclass(unsafe_hash=True)
|
|
class Model:
|
|
"""
|
|
Represents a machine learning model configuration.
|
|
|
|
Attributes:
|
|
name (str): Name of the model.
|
|
base_provider (str): Default provider for the model.
|
|
best_provider (ProviderType): The preferred provider for the model, typically with retry logic.
|
|
"""
|
|
name: str
|
|
base_provider: str
|
|
best_provider: ProviderType = None
|
|
|
|
@staticmethod
|
|
def __all__() -> list[str]:
|
|
"""Returns a list of all model names."""
|
|
return _all_models
|
|
|
|
default = Model(
|
|
name = "",
|
|
base_provider = "",
|
|
best_provider = RetryProvider([
|
|
Bing,
|
|
ChatgptAi, GptGo,
|
|
You,
|
|
Chatgpt4Online
|
|
])
|
|
)
|
|
|
|
# GPT-3.5 too, but all providers supports long requests and responses
|
|
gpt_35_long = Model(
|
|
name = 'gpt-3.5-turbo',
|
|
base_provider = 'openai',
|
|
best_provider = RetryProvider([
|
|
FreeGpt, You,
|
|
Chatgpt4Online,
|
|
ChatgptNext,
|
|
ChatgptDemo,
|
|
Gpt6,
|
|
])
|
|
)
|
|
|
|
# GPT-3.5 / GPT-4
|
|
gpt_35_turbo = Model(
|
|
name = 'gpt-3.5-turbo',
|
|
base_provider = 'openai',
|
|
best_provider = RetryProvider([
|
|
GptGo, You,
|
|
GptForLove, ChatBase,
|
|
Chatgpt4Online,
|
|
])
|
|
)
|
|
|
|
gpt_4 = Model(
|
|
name = 'gpt-4',
|
|
base_provider = 'openai',
|
|
best_provider = RetryProvider([
|
|
Bing, Liaobots,
|
|
])
|
|
)
|
|
|
|
gpt_4_turbo = Model(
|
|
name = 'gpt-4-turbo',
|
|
base_provider = 'openai',
|
|
best_provider = Bing
|
|
)
|
|
|
|
gigachat = Model(
|
|
name = 'GigaChat:latest',
|
|
base_provider = 'gigachat',
|
|
best_provider = GigaChat
|
|
)
|
|
|
|
gigachat_plus = Model(
|
|
name = 'GigaChat-Plus',
|
|
base_provider = 'gigachat',
|
|
best_provider = GigaChat
|
|
)
|
|
|
|
gigachat_pro = Model(
|
|
name = 'GigaChat-Pro',
|
|
base_provider = 'gigachat',
|
|
best_provider = GigaChat
|
|
)
|
|
|
|
llama2_7b = Model(
|
|
name = "meta-llama/Llama-2-7b-chat-hf",
|
|
base_provider = 'meta',
|
|
best_provider = RetryProvider([Llama2, DeepInfra])
|
|
)
|
|
|
|
llama2_13b = Model(
|
|
name = "meta-llama/Llama-2-13b-chat-hf",
|
|
base_provider = 'meta',
|
|
best_provider = RetryProvider([Llama2, DeepInfra])
|
|
)
|
|
|
|
llama2_70b = Model(
|
|
name = "meta-llama/Llama-2-70b-chat-hf",
|
|
base_provider = "meta",
|
|
best_provider = RetryProvider([Llama2, DeepInfra, HuggingChat])
|
|
)
|
|
|
|
codellama_34b_instruct = Model(
|
|
name = "codellama/CodeLlama-34b-Instruct-hf",
|
|
base_provider = "meta",
|
|
best_provider = RetryProvider([HuggingChat, DeepInfra])
|
|
)
|
|
|
|
codellama_70b_instruct = Model(
|
|
name = "codellama/CodeLlama-70b-Instruct-hf",
|
|
base_provider = "meta",
|
|
best_provider = RetryProvider([DeepInfra, PerplexityLabs])
|
|
)
|
|
|
|
# Mistral
|
|
mixtral_8x7b = Model(
|
|
name = "mistralai/Mixtral-8x7B-Instruct-v0.1",
|
|
base_provider = "huggingface",
|
|
best_provider = RetryProvider([DeepInfra, HuggingChat, HuggingFace, PerplexityLabs])
|
|
)
|
|
|
|
mistral_7b = Model(
|
|
name = "mistralai/Mistral-7B-Instruct-v0.1",
|
|
base_provider = "huggingface",
|
|
best_provider = RetryProvider([DeepInfra, HuggingChat, HuggingFace, PerplexityLabs])
|
|
)
|
|
|
|
# Misc models
|
|
dolphin_mixtral_8x7b = Model(
|
|
name = "cognitivecomputations/dolphin-2.6-mixtral-8x7b",
|
|
base_provider = "huggingface",
|
|
best_provider = DeepInfra
|
|
)
|
|
|
|
lzlv_70b = Model(
|
|
name = "lizpreciatior/lzlv_70b_fp16_hf",
|
|
base_provider = "huggingface",
|
|
best_provider = DeepInfra
|
|
)
|
|
|
|
airoboros_70b = Model(
|
|
name = "deepinfra/airoboros-70b",
|
|
base_provider = "huggingface",
|
|
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(
|
|
name = "openchat/openchat_3.5",
|
|
base_provider = "huggingface",
|
|
best_provider = RetryProvider([DeepInfra, HuggingChat])
|
|
)
|
|
|
|
# Bard
|
|
gemini = bard = palm = Model(
|
|
name = 'gemini',
|
|
base_provider = 'google',
|
|
best_provider = Gemini
|
|
)
|
|
|
|
claude_v2 = Model(
|
|
name = 'claude-v2',
|
|
base_provider = 'anthropic',
|
|
best_provider = RetryProvider([Vercel])
|
|
)
|
|
|
|
claude_3_opus = Model(
|
|
name = 'claude-3-opus',
|
|
base_provider = 'anthropic',
|
|
best_provider = You
|
|
)
|
|
|
|
claude_3_sonnet = Model(
|
|
name = 'claude-3-sonnet',
|
|
base_provider = 'anthropic',
|
|
best_provider = You
|
|
)
|
|
|
|
gpt_35_turbo_16k = Model(
|
|
name = 'gpt-3.5-turbo-16k',
|
|
base_provider = 'openai',
|
|
best_provider = gpt_35_long.best_provider
|
|
)
|
|
|
|
gpt_35_turbo_16k_0613 = Model(
|
|
name = 'gpt-3.5-turbo-16k-0613',
|
|
base_provider = 'openai',
|
|
best_provider = gpt_35_long.best_provider
|
|
)
|
|
|
|
gpt_35_turbo_0613 = Model(
|
|
name = 'gpt-3.5-turbo-0613',
|
|
base_provider = 'openai',
|
|
best_provider = gpt_35_turbo.best_provider
|
|
)
|
|
|
|
gpt_4_0613 = Model(
|
|
name = 'gpt-4-0613',
|
|
base_provider = 'openai',
|
|
best_provider = gpt_4.best_provider
|
|
)
|
|
|
|
gpt_4_32k = Model(
|
|
name = 'gpt-4-32k',
|
|
base_provider = 'openai',
|
|
best_provider = gpt_4.best_provider
|
|
)
|
|
|
|
gpt_4_32k_0613 = Model(
|
|
name = 'gpt-4-32k-0613',
|
|
base_provider = 'openai',
|
|
best_provider = gpt_4.best_provider
|
|
)
|
|
|
|
gemini_pro = Model(
|
|
name = 'gemini-pro',
|
|
base_provider = 'google',
|
|
best_provider = RetryProvider([GeminiProChat, You])
|
|
)
|
|
|
|
pi = Model(
|
|
name = 'pi',
|
|
base_provider = 'inflection',
|
|
best_provider = Pi
|
|
)
|
|
|
|
class ModelUtils:
|
|
"""
|
|
Utility class for mapping string identifiers to Model instances.
|
|
|
|
Attributes:
|
|
convert (dict[str, Model]): Dictionary mapping model string identifiers to Model instances.
|
|
"""
|
|
convert: dict[str, Model] = {
|
|
# gpt-3.5
|
|
'gpt-3.5-turbo' : gpt_35_turbo,
|
|
'gpt-3.5-turbo-0613' : gpt_35_turbo_0613,
|
|
'gpt-3.5-turbo-16k' : gpt_35_turbo_16k,
|
|
'gpt-3.5-turbo-16k-0613' : gpt_35_turbo_16k_0613,
|
|
|
|
'gpt-3.5-long': gpt_35_long,
|
|
|
|
# gpt-4
|
|
'gpt-4' : gpt_4,
|
|
'gpt-4-0613' : gpt_4_0613,
|
|
'gpt-4-32k' : gpt_4_32k,
|
|
'gpt-4-32k-0613' : gpt_4_32k_0613,
|
|
'gpt-4-turbo' : gpt_4_turbo,
|
|
|
|
# Llama 2
|
|
'llama2-7b' : llama2_7b,
|
|
'llama2-13b': llama2_13b,
|
|
'llama2-70b': llama2_70b,
|
|
'codellama-34b-instruct': codellama_34b_instruct,
|
|
'codellama-70b-instruct': codellama_70b_instruct,
|
|
|
|
# GigaChat
|
|
'gigachat' : gigachat,
|
|
'gigachat_plus': gigachat_plus,
|
|
'gigachat_pro' : gigachat_pro,
|
|
|
|
'mixtral-8x7b': mixtral_8x7b,
|
|
'mistral-7b': mistral_7b,
|
|
'dolphin-mixtral-8x7b': dolphin_mixtral_8x7b,
|
|
'lzlv-70b': lzlv_70b,
|
|
'airoboros-70b': airoboros_70b,
|
|
'airoboros-l2-70b': airoboros_l2_70b,
|
|
'openchat_3.5': openchat_35,
|
|
'gemini': gemini,
|
|
'gemini-pro': gemini_pro,
|
|
'claude-v2': claude_v2,
|
|
'claude-3-opus': claude_3_opus,
|
|
'claude-3-sonnet': claude_3_sonnet,
|
|
'pi': pi
|
|
}
|
|
|
|
_all_models = list(ModelUtils.convert.keys())
|