Show HN: Chat with multiple LLMs: o1-high-effort, Sonnet 3.5, GPT-4o, and more • Hacker News Board

Re: Show HN: Chat with multiple LLMs: o1-high-effort, Sonnet 3.5, GPT-4o, and more

#21

Post by pogue » Fri, Jan 24, 2025, 3:43 AM UTC

Fyi: Duckduckgo has a free ai chat with o1-mini at https://duck.ai

They also offer Claude 3 Haiku, Llama 3.1 70B and Mixtral 8x7B. No account required and they say it's not logged + anonymous. I'm sure they have limits but I never reached them.

Re: Show HN: Chat with multiple LLMs: o1-high-effort, Sonnet 3.5, GPT-4o, and more

#22

Post by wslh » Fri, Jan 24, 2025, 3:47 AM UTC

We have Geppetto, a FOSS that you can ask questions within Slack and using several services. Feel free to fork it: https://github.com/Deeptechia/geppetto>

Re: Show HN: Chat with multiple LLMs: o1-high-effort, Sonnet 3.5, GPT-4o, and more

#23

Post by iandanforth » Fri, Jan 24, 2025, 5:08 AM UTC

oidar wrote: post #5

This is very good at this right now too: https://melty.sh/chorus

You should also note that Chorus allows you to bring your own API keys and interact with local models. Both features I very much appreciate!

Re: Show HN: Chat with multiple LLMs: o1-high-effort, Sonnet 3.5, GPT-4o, and more

#24

Post by cr4zy » Fri, Jan 24, 2025, 5:24 AM UTC

moralestapia wrote: post #20

Unlimited free OpenAI o1? OP, isn't that really expensive to maintain?

It's not unlimited free unfortunately. After some free use, we provide monthly usage-tier plans. But there are no rate limits like other providers as you can move up to the next usage-tier.

Re: Show HN: Chat with multiple LLMs: o1-high-effort, Sonnet 3.5, GPT-4o, and more

#25

Post by cr4zy » Fri, Jan 24, 2025, 5:30 AM UTC

So it looks like some folks are getting errors with the non-streaming models, i.e. the o1 models. I think their long running cxns with zero packets may cause some networks to drop the requests. Will look into a hearbeat/keepalive on those.

Re: Show HN: Chat with multiple LLMs: o1-high-effort, Sonnet 3.5, GPT-4o, and more

#26

Post by cr4zy » Fri, Jan 24, 2025, 5:46 AM UTC

iandanforth wrote: post #23

oidar wrote: post #5

This is very good at this right now too: https://melty.sh/chorus

You should also note that Chorus allows you to bring your own API keys and interact with local models. Both features I very much appreciate!

One tradeoff of bringing your own API keys is that as you add more model providers, you get more billing accounts to deal with. Chorus also doesn't have an incentive to efficiently use your tokens. We save 67% on Anthropic token costs using Claude Caching. We also use cheaper "task models" for conversation titles, tagging, and parts of the RAG pipeline which all drastically cuts token costs.

For local models I highly recommend https://github.com/crizCraig/open-webui

They do the side by side thing that Chorus does and you can serve it to anywhere including your phone.

Re: Show HN: Chat with multiple LLMs: o1-high-effort, Sonnet 3.5, GPT-4o, and more

#27

Post by yakorevivan » Fri, Jan 24, 2025, 7:49 AM UTC

[dead]

Re: Show HN: Chat with multiple LLMs: o1-high-effort, Sonnet 3.5, GPT-4o, and more

#28

Post by ilrwbwrkhv » Fri, Jan 24, 2025, 8:15 AM UTC

If it doesn't have DeepSeek R1 it is a non starter.

Re: Show HN: Chat with multiple LLMs: o1-high-effort, Sonnet 3.5, GPT-4o, and more

#29

Post by rany_ » Fri, Jan 24, 2025, 9:05 AM UTC

pogue wrote: post #21

Fyi: Duckduckgo has a free ai chat with o1-mini at https://duck.ai They also offer Claude 3 Haiku, Llama 3.1 70B and Mixtral 8x7B. No account required and they say it's not logged + anonymous. I'm sure they have limits but I never reached them.

They offer 4o-mini, not o1-mini.