God Mode Pod
Region: BE
Monday 13 July 2026 15:01:38 GMT
22043
880
58
111
Music
Download
Comments
Charles A. Lee :
i think these companies just want cheaper models bruh. the american AI models need to make their models cheaper and better if they want to stay on top of the competition.
2026-07-17 06:51:43
0
gra_v_t :
Started with deepseek and never left
2026-07-13 20:49:33
43
Michael Burton :
do china has conquered america
2026-07-16 15:15:58
0
Joey :
fable for implementation plan, deepseek agents for implementation, fable to check the work and bugfix. seems to work ok
2026-07-14 09:32:27
15
Jouunaas :
which china model is the best for coding on vscode?
2026-07-14 06:21:36
0
houariZEN :
only kimi is OK to code but claude and codex a still way better I've tried all of them on his videos
2026-07-14 12:28:35
6
Botanical Gardens :
My parents always told me in life, you get what you pay for.
2026-07-16 12:56:49
0
Melonading :
fr Ive been using qwen3.6:35b MOE model and its been great, especially when you connect context7, sequential thinking and seargnx its insanely good to run offline
2026-07-14 21:10:37
1
Chob :
Different use cases, if you do coding for your own project, choose the best and affordable model for you. But if you provide SAAS service for your customers, something like Claude is definitely an overkill and expensive as fuck, in the end most companies will choose qwen or deepseek, EU more likely qwen, because they have EU server
2026-07-14 15:37:38
0
Markus99 :
Anthropic waited too long to release Fable and for the IPO. 🤔
2026-07-13 18:50:29
6
thomas.b :
Misleading to say cursor uses Kimi just because it's the foundation of Composer as if they didn't fine-tune the hell out of it
2026-07-14 15:29:17
0
John anton :
me likey deepseek and kimi
2026-07-14 16:05:02
1
Ahmed Shifau :
you dont need the big models, for everyday use. a good harness is much more important. for everyday use open webui + qwen3.5 (7b) is fine as a normal chat for answering. you can also use API to link to deepseek or qwen and change model as needed.
2026-07-14 16:42:31
0
rodney.j.thomas :
Claude can counter by optimizing compute and token usage with integration of tools like graphify? Reduce token costs for users and compute costs for themselves?
2026-07-15 06:58:03
0
Kwok :
I use glm 5.2
2026-07-15 17:32:53
0
祖龍 :
昂贵的 API 真的没必要，因为这个昂贵的只是电力，在电力对中国来说很容易获得。
2026-07-18 06:54:09
0
Snow Ball :
Oh lawdy lawdy waa is Trumpet gonna do about this national security threat?
2026-07-14 12:44:23
1
fragola_jam :
Do we rate the Chinese models? I asked kimi
2026-07-13 16:59:17
0
Berliner Scharfe :
look at nvidias nim apis and models 😉
2026-07-13 20:07:45
0
A SaU :
🥰🥰🥰
2026-07-18 07:46:22
0
Zagros :
❤️❤️❤️
2026-07-18 11:10:08
0
rickeyhendersons :
That will be the downfall of the us ai industry. No one wants to pay a us company similar with how gay apple ceo cook sourcing everything from China
2026-07-16 09:55:40
0
Lorris :
but you aren't a big company, running those dumb Local AI require real technical skills that you don't need as much with Claude or chatgpt. And running those local AI require some monstrous vram that you don't have or very few have. Local AI or even using those Chinese AI in cloud server is more expensive than Claude or Chatgpt or the new Grok model. Local AI Chineseis great for big companies that need total freedom, but that freedom has a cost. And those big companies probably use both Chinese AI and USA model, they would lose too much time otherwise. They use the smarter model to steer the dumb one.
2026-07-14 17:19:09
0
To see more videos from user @godmodepod, please go to the Tikwm
homepage.
