search

xtekky / gpt4free

The official gpt4free repository | various collection of powerful language models
https://g4f.ai
GNU General Public License v3.0
60.05k stars 13.22k forks source link

Advertisement in answers #2233

Open nBalykin opened 2 days ago

nBalykin commented 2 days ago

I use llama-3.1-70b and sometimes I get this answer. But gpt-4o has this issue too:

One message exceeds the 1000chars per message limit. Join our discord for more: [https://discord.com/invite/q55gsH8z5F](https://discord.com/invite/q55gsH8z5F)

I use g4f as telegram bot. My request code (python)

response = g4f.ChatCompletion.create(
model = "llama-3.1-70b",
messages = text
)

Screenshot_2024-09-19-19-26-53-75_4495e6112227b794374c62341ece5829

TheFirstNoob commented 2 days ago

Try to use other provider for this model like HuggingChat. I think its provider problem.

response = g4f.ChatCompletion.create(
model = "llama-3.1-70b",
provider="HuggingChat",
messages = text
)
nBalykin commented 1 day ago

Upd: I've removed Airforce in section llama in models.py file. Now I don't have this message. I think this is Airforce provider advertisement.

187 llama_3_1_70b = Model(
188     name          = "llama-3.1-70b",
189     base_provider = "Meta",
190     best_provider = IterListProvider([DDG, HuggingChat, FreeGpt, Blackbox, TeachAnything, Free2GPT, HuggingFace, PerplexityLabs])