#UnrealEngine5 #Twitch #Crypto degen. 4th bear cycle survivor. https://teapaw.org/t.co/kUjPFyO3sR platform creator, catch up on $KAS news and updates, join the public forums!
MOM THE MEATLOAF
Joined July 2009
- Tweets11.3K
- Following232
- Followers1.1K
- Likes11.6K
Pinned Tweet
Ever played the legendary Heroes of Might and Magic series? Pick up the just released Olden Era title Steam key with 25% discount, available only for the next 2 weeks to celebrate game's Early Access launch on: nexus.gg/mmostars/heroes-of-β¦
π€ Paid partnership
1
1
555
Tested Qoduck thinking it was gonan be Qwen4 27b small model, turns out it's Qwen4-Max-Preview complete disappointment for the masses inc. There was no feel to any of the test runs that it would be even above Deepseek 4.1 flash level. π
155
Unreal Engine 5.9 Preview run with Hy4 from Tencent LLM using the new WASM bundled package and python bridge for web export. You can change anything now in Unreal with all of the best stuff like Niagara/lighting powering it and have the same gameplay render natively in WebGPU.
1
1
161
Fable 5.2 tests video, doesn't seem to be worth the 2x the price over Opus 5.5 and Opus already solved web dev completely for a paid model. youtube.com/live/QwaU4yU0PdQ
1
162
So trash get's hacked, kaspa:native completely not effected in any way, but somehow during the hack, we get shorted down by 200mil paper KAS. Make it make sense. This cex should go down like FTX at this point, only proper outcome for them.
1
2
225
We had DirectX 2 3 5 6 7 8 9 10 11 then 12. It has been 11 years since the last one, what happened chat?
1
87
Didn't test tool call ability in local app, but for chatbot level insanely strong model for 2b size teapaw.org/liquidai/status/211512β¦
Introducing π»π΅πΌπΉ: our new generation of Liquid Foundation Models. With a new tri-hybrid architecture and new multimodal encoders, the LFM3 family is the ideal model suite for specialized intelligence.
> 6 model instances, across text, audio, vision, and omni modalities, generative and decision models.
> π»π΅πΌπΉ-π·.π½π±: text, 256k context window.
> π»π΅πΌπΉ-π
π»-πΈπ±: vision-language model with video-input support for spatial understanding
> π»π΅πΌπΉ-π°ππππ-πΈπ±: tool calls, voice chat, reasoning in 18 languages.
> π»π΅πΌπΉ-πΎπππ-πΈπ±: an omni model combines vision, audio, and language.
> ππ·-πΈπ± and ππ·-πΎπππ-πΈπ±: decision models with typed outputs and calibrated probabilities in a single pass. Omni adds audio input.
> fine-tune on 1 GPU.
> a new tri-hybrid architecture
> architecture optimized for 14 processor classes across GPUs, CPUs, and NPUs.
> π»π΅πΌπΉ-π
π»-πΈπ± is a beast. It performs competitively with Gemma 4 12B (6x larger), and Gemini 3.5 Flash Lite, and GPT-5.6 Luna, with 1.9B parameters.
> π»π΅πΌπΉ-π·.π½π± is #1 on the Pareto frontier of device intelligence πππππππ.ππππππ.ππ
Every modality for specialized intelligence. Pick the one that fits your use case, customize it, and run it on-device.
1/ π§΅
2
51
Do not recommend buying kimi sub, the subsidized usage amount is so small that it being 2x cheaper then opus, will give you even less use then opus overall.
42
teapaw.org/cerebras/status/211256β¦ Watching the Cerebras comments section is gold, sucks that tradefi is pulling money out of crypto when hamsters only care what to new stuff to gamble on next this days, now aping day1 IPOs thinking of the next dotcom run for AI.
46,225 mmΒ² of silicon
4 trillion transistors
900,000 AI-optimized cores
250 petaflops of AI compute
At the center of every CS-4 Wafer-Scale Backpack is the WSE-3 Turbo ‡οΈ
103
GPT Image 2.5 Sunburn is free on arena AI under direct use, if you have some image gen work in mind, do it now while free usage lasts.
54
