chat-with-gpt
human-eval
chat-with-gpt | human-eval | |
---|---|---|
39 | 6 | |
2,267 | 2,014 | |
- | 7.4% | |
5.3 | 0.0 | |
24 days ago | 3 months ago | |
TypeScript | Python | |
MIT License | MIT License |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
chat-with-gpt
- Different chatGPT 4 API Interface?
-
Using API as a more powerful ChatGPT?
you can use https://www.chatwithgpt.ai/ , I like this + it's open source.
- I use chatGPT for hours everyday and can say 100% it's been nerfed over the last month or so. As an example it can't solve the same types of css problems that it could before. Imagine if you were talking to someone everyday and their iq suddenly dropped 20%, you'd notice. People are noticing.
- Hot Take: ChatGPT is not getting dumber, you are.
-
Show HN: LLaMA tokenizer that runs in browser
https://github.com/cogentapps/chat-with-gpt/blob/main/app/sr...
-
Is there a way to use one conversation as context for another?
Do you have api access? I use this repo called chat with chatgpt https://github.com/cogentapps/chat-with-gpt It literally is chatgpt website same look feel and conversation history, but you can adjust the creativeness (called temperature) the system prompt and what version you want to use.
-
OpenAI Employee: GPT-4 hasn't gotten worse since March
We've been using https://github.com/cogentapps/chat-with-gpt for a while, and overall happy with it, but its development is mostly dead.
-
I have been talking to the GPT-4 32k context length version for a couple of hours straight, and I can see how the average person might mistake it for some kind of intelligent living being after a few hours.
There are apps available that will take care of all this for you: https://github.com/cogentapps/chat-with-gpt is a popular one, if you need help to get it up and running just ask GPT-4 with 'browse with bing' to take a look and lead you by the hand
- Has ChatGPT Been Neutered?
-
What's the best ChatGPT web or Mac desktop client that let's you organize or search your past conversations?
So far I've tried https://github.com/ztjhz/BetterChatGPT and https://github.com/cogentapps/chat-with-gpt but I'm wondering if there is anything out there more robust? A Mac desktop client would be preferred if anyone knows? I tried MacGPT but the main functionality I am looking for is an ability to sort, search and potentially tag my conversations. Any leads would be appreciated! Much thanks!
human-eval
-
Nathan Lambert review of LLAMA 2: Open-Source LLM from Meta
Code / math / reasoning: Not much discussion of code data in the paper and RLHF process. For instance, StarCoder at 15 billion parameters beats the best model at 40.8 for HumanEval and 49.5 MBPP (Python).
-
New ChatGPT rival, Claude 2, launches for open beta testing
In terms of coding capabilities, Claude 2 demonstrated a reported increase in proficiency. Its score on the Codex HumanEval, a Python programming test, rose from 56 percent to 71.2 percent. Similarly, on GSM8k, a test comprising grade-school math problems, it improved from 85.2 to 88 percent.
-
Hot Take: ChatGPT is not getting dumber, you are.
there are for code https://github.com/openai/human-eval
-
Official WizardCoder-15B-V1.0 Released! Can Achieve 59.8% Pass@1 on HumanEval!
❗Note: The above table conducts a comprehensive comparison of our WizardCoder with other models on the HumanEval and MBPP benchmarks. We adhere to the approach outlined in previous studies by generating 20 samples for each problem to estimate the pass@1 score and evaluate it with the same code. The scores of GPT4 and GPT3.5 reported by OpenAI are 67.0 and 48.1 (maybe these are the early version of GPT4&3.5).
- A new way to predict when software jobs will become automated
- OpenAI Codex - The Model behind GitHub Copilot
What are some alternatives?
ChatGPT-Next-Web - A cross-platform ChatGPT/Gemini UI (Web / PWA / Linux / Win / MacOS). 一键拥有你自己的跨平台 ChatGPT/Gemini 应用。
evalplus - EvalPlus for rigourous evaluation of LLM-synthesized code
LibreChat - Enhanced ChatGPT Clone: Features OpenAI, Assistants API, Azure, Groq, GPT-4 Vision, Mistral, Bing, Anthropic, OpenRouter, Vertex AI, Gemini, AI model switching, message search, langchain, DALL-E-3, ChatGPT Plugins, OpenAI Functions, Secure Multi-User System, Presets, completely open-source for self-hosting. More features in development
llm-humaneval-benchmarks
chatwithme.chat - Open source ChatGPT UI
ggml - Tensor library for machine learning
plurality - A cornucopia of open source UIs built with ChatGPT API.
WizardLM - Family of instruction-following LLMs powered by Evol-Instruct: WizardLM, WizardCoder and WizardMath
ruby-openai - OpenAI API + Ruby! 🤖❤️ Now with Assistants v2, Batches & Ollama/Groq 🚀
llama - Inference code for Llama models
vocode-python - 🤖 Build voice-based LLM agents. Modular + open source.
chatgpt-web - 用 Express 和 Vue3 搭建的 ChatGPT 演示网页