@UseAllOverTools @ollama What hardware will run that? Will my 64GB Mac Studio run it?
English
Roger
5.5K posts
Roger
@roger
Family man, Fractional software leader, 1 x Exit, Rubyist, founder of @openrangedevs + @west17media.
Calgary, Alberta Beigetreten Haziran 2011
486 Folgt959 Follower
@ollama ollama cloud for deepseek? why bother when you can run these giants locally already. table says it all.
English
🐳Ollama is working to have DeepSeek-V4-Pro and Flash available on Ollama's cloud.
DeepSeek@deepseek_ai
🚀 DeepSeek-V4 Preview is officially live & open-sourced! Welcome to the era of cost-effective 1M context length. 🔹 DeepSeek-V4-Pro: 1.6T total / 49B active params. Performance rivaling the world's top closed-source models. 🔹 DeepSeek-V4-Flash: 284B total / 13B active params. Your fast, efficient, and economical choice. Try it now at chat.deepseek.com via Expert Mode / Instant Mode. API is updated & available today! 📄 Tech Report: huggingface.co/deepseek-ai/De… 🤗 Open Weights: huggingface.co/collections/de… 1/n
English
@deepseek_ai @grok could Deepseek V4 Flash run on a Dell GB10 with 128GB of Memory or better yet a Mac Studio with 64GB of Unified Memory?
English
🚀 DeepSeek-V4 Preview is officially live & open-sourced! Welcome to the era of cost-effective 1M context length.
🔹 DeepSeek-V4-Pro: 1.6T total / 49B active params. Performance rivaling the world's top closed-source models.
🔹 DeepSeek-V4-Flash: 284B total / 13B active params. Your fast, efficient, and economical choice.
Try it now at chat.deepseek.com via Expert Mode / Instant Mode. API is updated & available today!
📄 Tech Report: huggingface.co/deepseek-ai/De…
🤗 Open Weights: huggingface.co/collections/de…
1/n
English
I wonder if a DGX can run the Flash model and perform adequately. The Active Parameters makes me think it’s possible. I’m far from an expert on this, so I could be way off.
DeepSeek@deepseek_ai
DeepSeek-V4-Flash 🔹 Reasoning capabilities closely approach V4-Pro. 🔹 Performs on par with V4-Pro on simple Agent tasks. 🔹 Smaller parameter size, faster response times, and highly cost-effective API pricing. 3/n
English
English
@inventur_es @REIGNSBACH Miss that; it was literally the most fun drinking I’ve ever had. So many memories and cool interactions.
English
@REIGNSBACH We call them Drinkable streets. Very common in the UK
English
Do ppl in london just stand around in the middle of the road while drinking and chatting??? Genuinely asking here
peter claffey updates@pclaffeycentral
okay..
English
GPT 5.5 benchmarks
BUT w/o hiding the ones where they lose to Opus 4.7 lol
also check vs Mythos
marketing bastards got me again
OpenAI@OpenAI
GPT-5.5 delivers this step up in intelligence without compromising on speed. GPT-5.5 matches GPT-5.4 per-token latency in real-world serving, while performing better across nearly every evaluation we measured. It also uses significantly fewer tokens to complete the same Codex tasks, making it more efficient as well as more capable.
English
@dalethomas50 @NousResearch Without Hermes Agent I’m not sure, but with Hermes Agent there are tools and other optimizations. They are trying to turn the platform into the optimal leverage for the Hermes Agent. It’s an interesting model.
English
@NousResearch I did sign up for the $10 plan just to play with it. But is there any advantage of going through you vs. OpenRouter?
English
New subscription tiers are live on Nous Portal
→ Plus ($20)
→ Super ($100)
→ Ultra ($200)
Bonus credits on signups, upgrades, and renewals: +$2 on Plus / +$10 on Super / +$20 on Ultra
All tiers include access to:
→ 300+ models
→ Bundled tool usage
portal.nousresearch.com/manage-subscri…
English
@ChrisLikesAI @NousResearch I was looking too! Thanks for asking and answering gang!
English
@NousResearch @NousResearch are the 300+ models listed anywhere? My Claude expires 4/25 :) Perfect timing.
English
New best local model for y'all
16GB-64GB rejoice, the chosen one has arrived.
huggingface.co/Qwen/Qwen3.6-2…
English
@_MaxBlade People use opus4.7 because they can use it off the shelves and not k2.6 because it’s a nightmare for nontechnical people to set it up.
English
Everyone says Hermes / Openclaw is unusable without Opus 4.7
I just ran two Hermes agents head to head, one with k2.6 and one with opus 4.7, through three heavily agentic tasks with one prompt.
OPUS DID NOT COMPLETE 😳
Kimi completed all tasks.
It took 4x longer, but is 10x less expensive to run.
This is SHOCKING.
English
X getting rid of communities is a friendly reminder to never build your main thing on someone else's platform.
English
Do what Klein did for Oil Sands and if there is an opportunity money will follow.
Marc Nixon@MarcNixon24
Steven McKinnon announces Canada wants to take on Elon Musk and launch rockets into space This could be a $40 billion industry What’s your thought? Good or bad investment?
English
@CodeWithZeee @TheGeorgePu Actually I downgraded to Pro for a few months when I was focused on other things that didn’t require opus, nor a ton of Sonnet.
English
@TheGeorgePu I’d imagine it’s a very small number of pro using Claude code
English
Scaling up Kimi K2.6 with more NVIDIA Blackwell GPUs! Will be adding more in the coming days.
Try it with OpenClaw:
ollama launch openclaw --model kimi-k2.6:cloud
Try it with Hermes Agent:
ollama launch hermes --model kimi-k2.6:cloud
Try it with Claude Code:
ollama launch claude --model kimi-k2.6:cloud
Chat with the model:
ollama run kimi-k2.6:cloud
English
@mojito_flow @venturetwins I agree with you Verdis, children don’t need AI.
English
@venturetwins I’m very pro-AI but this parent is right. Children should not be exposed to LLMs while their brains are maturing. This will be worse than LLM psychosis.
I’m glad I didn’t have access to LLM’s growing up.
English
