NEW DeepSeek-V3 is INSANE (FREE): RIP 3.5 Sonnet & O1?

[@JulianGoldieSEO]

🚀 Get a FREE SEO strategy Session + Discount Now: go.juliangoldie.com/strategy-session - Signup for 3 months and get 1 month FREE - Signup for 6 months, get 2 months for FREE! - Signup for 12 months, get 4 months for FREE! 🤯 Want more money, traffic and sales from SEO? Join the SEO Elite Circle: - go.juliangoldie.com/buy-mastermind - BLACK FRIDAY DEAL! Save $2,600 with the code ELITE Click below for FREE access to ✅ 50 FREE AI SEO TOOLS 🔥 200+ AI SEO Prompts! 📈 FREE AI SEO COMMUNITY with 2,000 SEOs ! 🚀 Free AI SEO Course 🏆 Plus TODAY's Video NOTES... go.juliangoldie.com/chat-gpt-prompts

[@SpeedToLead-io]

When I ask DeepSeek, "What version are you?" it responds, "I'm ChatGPT-4, an AI language model developed by OpenAI."

[@Anders01]

I got this reply from DeepSeek: "I'm DeepSeek-V3, an AI created exclusively by the Chinese Company DeepSeek. I'll do my best to help you."

[@philpilkington]

I found the same with a version of Qwen. Is this proof of anything? Don't know anything of model training from scratch

[@PJ-qy9xg]

Doesn't surprise me China steals everything

[@unlimitedbytes-random]

The model absolutely doesn't perform. Claude gives way better result in every test than that model.

[@huk2617]

Cope more

[@Niranjanjiragale-m9p]

Amazing tool

[@Copa20777]

China is cooking 😅

[@elawchess]

At 0:57 I think you mispoke. You say Deepseek is outperforming O1 while pointing to GPT-4o in the diagram

[@shoufen]

I can’t find deepseek v 3. I only see v 2.5.

[@YoucefKrm-e5m]

it's the same bro

[@CodahWasTaken]

Short resume: *It sucks.*

[@huk2617]

Cope more

[@Fwuzeem]

I've used it a lot today, and I still prefer Claude. It made some really basic mistakes

[@JoshDingus]

You seo link free training is broke

[@johnfleming1229]

What about the fact that this is Chinese control, no concerns?

[@Atheist-Libertarian]

It's open source right? So you can download it and run it on your GPUs

[@huk2617]

Not really, it is open source after all, unlike the wealth of closed-source U.S. models

[@blengi]

how does it do on ARC and frontier math compared to o3?

[@BACA01]

It's a rebadged GPT4

[@pondeify]

it literally told me it's ChatGPT when I asked what version it was...

[@adamgalas]

@@pondeify same here

[@philpilkington]

Ummm how did someone get lost since it's not released

[@Rami_Elkady]

Somalia also launched their own LLM model that outperforms Claude ... It is called BullShitLM ... You guys should not try it 😁

[@huk2617]

Cry more

[@mrki731]

China? No way.

[@andyatmosphere]

Once I heard china.... I was out.

[@sizwemsomi239]

You should throw away your phone to...

[@huk2617]

Cry more

[@tanakaaiko-]

Farewell haha😆😆