๐ We just supported the latest O1 models from @OpenAI Things to watch out for now:
- No system prompt support
- No streaming support, tool usage, batch calls or image inputs either.
- Temperature has to be set to 1 and you also need a larger max tokens for reasoning.
๐ป The new `gpt-4o-2024-08-06` is now available. Same performance with 50% lower input and 33% lower output cost.
Try it now: https://t.co/eDEsxgXyln
Exciting News from Chatbot Arena!
@GoogleDeepMind's new Gemini 1.5 Pro (Experimental 0801) has been tested in Arena for the past week, gathering over 12K community votes.
For the first time, Google Gemini has claimed the #1 spot, surpassing GPT-4o/Claude-3.5 with an impressive score of 1300 (!), and also achieving #1 on our Vision Leaderboard.
Gemini 1.5 Pro (0801) excels in multi-lingual tasks and delivers robust performance in technical areas like Math, Hard Prompts, and Coding.
Huge congrats to @GoogleDeepMind on this remarkable milestone!
Gemini (0801) Category Rankings:
- Overall: #1
- Math: #1-3
- Instruction-Following: #1-2
- Coding: #3-5
- Hard Prompts (English): #2-5
Come try the model and let us know your feedback!
More analysis below๐
๐ Just added several new models:
- llama-3-8b-instruct
- llama-3-70b-instruct
- mistral-7b-instruct
- mixtral-8x7b
- mixtral-8x22b
- mixtral-8x22b-instruct
Shoutout to @openrouter which empowers us to integrate new open-source models easily, we will continue to add more based on customer feedbacks.
You can now manage and test your prompts with all these fresh models!
Try them now: https://t.co/5e9QC7bsgQ
@Trace_Cohen My couch. MacBook Pro. Runpod for compute. Axolotl for training. OpenAI Playground and Anthropic Console for prompt engineering. Sometimes PromptKnit.
๐ Gemini 1.5 Pro is now available in all prompt editors and playgrounds.
It has 1M context window and 8k max output tokens. Definitely one of the best models so far.
Try it now: https://t.co/eDEsxgXyln