暴雨的節奏 
Pricing
Fine-tuning costs are broken down into two buckets: the initial training cost and usage cost:
Training: $0.008 / 1K Tokens
Usage input: $0.012 / 1K Tokens
Usage output: $0.016 / 1K Tokens
For example, a gpt-3.5-turbo fine-tuning job with a training file of 100,000 tokens that is trained for 3 epochs would have an expected cost of $2.40.
我後來都直接用 NextDNS 了，相對方便，缺點是只能用官方審核過的 hosts。
openrouter.ai 這服務對開發大型語言模型應用的團隊帶來極大的便利性，目前主流 LLM 的 API 造訪權限都要特別申請 eg. OpenAI GPT-4-32k, Anthropic 的 Claude, Google 的 Google: PaLM 2。就算是開源模型 Meta 的 LLAMA 2 雖然可以自己架設，但沒有好的硬體一樣沒用。openrouter.ai 解決了這個痛點，只要透過 openrouter.ai 提供的 API ，就可以調用主流的大型語言模型來加速開發，讓你的應用可以快速的測試在不同語言模型下的表現。
The services provided by "openrouter.ai" bring great convenience to teams developing large language model applications. Currently, the access rights to mainstream LLM APIs, such as OpenAI's GPT-4-32k, Anthropic's Claude, and Google's PaLM 2, require special applications. Even for open-source models like Meta's LLAMA 2, they are of no use without good hardware despite the possibility of self-hosting. openrouter.ai addresses this pain point. By using the API provided by openrouter.ai, you can call upon mainstream large language models to accelerate development, allowing your application to quickly test its performance under different language models.
#LLM #AI #GPT
Azure 大放送？直接一次開發 GPT-4 和 GPT-4-32k
基於 Nostr protocol 的短網址服務
最近午後都會有雷陣雨，下過之後的氣溫正常多了。 https://nostrcheck.me/media/public/nostrcheck.me_8059313661818768301691924754.webp
看得懂的應該都上年紀了，小時候還會拿鉛筆轉啊轉倒帶。 
原本以為睏霸數錢，結果天崩地裂 