6 ms·
"easily train cheaper OSS models" That's the claim, but I don't see it. All these open models I tested are WAY worse than GPT-3. (2?)
by singularity2001 4y ago
"easily train cheaper OSS models"
That's the claim, but I don't see it. All these open models I tested are WAY worse than GPT-3. (2?)
- arthur_sav 4y agoCan confirm, at least on Gpt4all. Just tried it and it's nothing like ChatGPT
- PartiallyTyped 4y agoChatGPT is GPT3.5(++)
- adlpz 4y agoIf you pay you can use GPT-4 within the ChatGPT UI.
- PartiallyTyped 4y agoI am aware, i was pointing out that chatgpt is not gpt 3.0 based but 3.5 and subsequent releases are further along.
- wsgeorge 4y agoDo you mean ChatGPT (on GPT-3.5 Turbo)? The foundation models on which they're built are only GPT-3 capable, and what most many people run locally are the lightest weight, quantized weights, so their performance is even more degraded. LLaMA-based models are popular because LLaMA beat GPT-3 on benchmarks. But ChatGPT runs on 3.5 Turbo and later, beating them. There's currently no open model that compares.