Every post of theirs in the library
Loading their posts
Lisan al Gaib
Writes about AI. 91% of posts open with an observation. A typical post reaches 320 likes.
- Posts
- 43
- Likes
- 23K
- Typical
- 320
- Best
- 3K
What they write about
How they open
Every post of theirs in the library
| Post | Topic | Likes | Replies | Reposts | Score | Posted | Actions |
|---|---|---|---|---|---|---|---|
| @scaling01guys can we please slow down. we are at the beginning of march and: - xAI imploded - Qwen … | AI | 3K | 85 | — | 54 | ||
| @scaling01> openai > non-profit > for the benefit of all of humanity > begs government to secure the… | AI | 2.4K | 61 | — | 52 | ||
| @scaling01GPT-5.5 is on par with Claude Mythos - GPT-5.5 average pass rate of 71.4% (±8.0%) - Mythos… | AI | 1.7K | 82 | — | 50 | ||
| @scaling01at this point you have to believe in one of three things: - ZAI is benchmaxxing - ZAI foun… | AI | 1.5K | 147 | — | 54 | ||
| @scaling01Its so over for OpenAI and Anthropic. Gemini 3 Pro Benchmarks 37.5% on HLE 31.1% on ARC-AG… | AI | 1.5K | 103 | — | 49 | ||
| @scaling01it's incredible how these models are now narrowly super intelligent at discrete mathematic… | AI | 907 | 58 | — | 45 | ||
| @scaling01Anthropic is overtaking OpenAI. Their revenue has more than doubled in 3 months! Anthropic… | AI | 800 | 30 | — | 43 | ||
| @scaling01The new version completely smashes GPT-5.5 and the previous Mythos version. Before Mythos … | AI | 732 | 27 | — | 42 | ||
| @scaling01GPT-6 spotted in Chatgpt GPT-6-Astra GPT-6-Astra-WM GPT-6-Pro WM might be work-mode or wor… | AI | 629 | 12 | — | 47 | ||
| @scaling01first 5T, now it's 10T if this is true and they finish training a 10T model this year, the… | AI | 579 | 36 | — | 44 | ||
| @scaling01Grok-5 started training and is now officially postponed to Q1 2026. Musk also said it will… | AI | 557 | 49 | — | 41 | ||
| @scaling01The Fable 5 relaunch is kind of fake: "some routine tasks like coding and debugging will f… | AI | 484 | 42 | — | 40 | ||
| @scaling01I have no doubt in my mind anymore that Mythos Preview is their massive ~10T teacher model… | AI | 461 | 12 | — | 41 | ||
| @scaling01Qwen3.8-Max is now available and weights will be released next week it's a 2.4T@95B MoE mo… | AI | 445 | 28 | — | 39 | ||
| @scaling01The compute story is cope. The gatekeeping story is cope. Mythos is genuinely much stronge… | AI | 404 | 24 | — | 38 | ||
| @scaling01Meta might actually be back with Muse Spark Still behind OpenAI, Anthropic and Google, but… | AI | 397 | 23 | — | 37 | ||
| @scaling01SpaceX acquired Cursor I expect SpaceX AI to be between Google and OpenAI by end of year C… | AI | 386 | 34 | — | 38 | ||
| @scaling01How naive are people? OpenAI is going to release GPT-5.5 very soon and it will be in the s… | AI | 367 | 40 | — | 38 | ||
| @scaling01"Kimi K3 performs significantly below the most recent frontier cyber-capable models" On UK… | AI | 362 | 33 | — | 37 | ||
| @scaling01if they make the same change for the normal subscription users too, then it's honestly ove… | SaaS | 329 | 33 | — | 37 | ||
| @scaling01Gemini 3.6 Flash is slightly more token-efficient than 3.5 but idk why you would take anyt… | AI | 324 | 24 | — | 36 | ||
| @scaling01LLMs might be stupid after all they can refuse to answer questions on which they are uncer… | AI | 320 | 21 | — | 36 | ||
| @scaling01foreign Anthropic employees saved "a license will no longer be required to export, reexpor… | AI | 314 | 8 | — | 35 | ||
| @scaling01honestly incredible considering its size much smaller than DeepSeek, Kimi and closed model… | AI | 299 | 14 | — | 35 | ||
| @scaling01ox-alpha is GLM-5.3-Flash with 320B@18B params - it beats the larger GLM-5.2 - it was pre-… | AI | 293 | 11 | — | 37 | ||
| @scaling01Gemini 3 Pro Preview now on aistudio Pricing: <=200K tokens • Input: $2.00 / Output: $12.0… | AI | 272 | 26 | — | 35 | ||
| @scaling01META could have trained DeepSeek-V3 at least 15 times using the compute budget of the Llam… | AI | 264 | 12 | — | 31 | ||
| @scaling01most interesting features in the Anthropic CC repo: - Kairos: always-on autonomous agent m… | AI | 256 | 6 | — | 33 | ||
| @scaling01Gemini 3.0 Pro made GPT-5.1 obsolete overnight, it's cheaper and much stronger Gemini 3.0 … | AI | 255 | 5 | — | 33 | ||
| @scaling01OpenAI and Microsoft changed their partnership but I think it comes down to this: "OpenAI … | AI | 249 | 19 | — | 34 | ||
| @scaling01OpenAI is no longer has a capped-profit structure they are now a PBC without a 100x ROI ca… | AI | 244 | 23 | — | 34 | ||
| @scaling01no surprises here the open-source panicans were just flapping their wings again another hi… | AI | 243 | 27 | — | 34 | ||
| @scaling01this sounds a bit more positive for Anthropic, but you never know with Trump if Mythos/Fab… | AI | 221 | 13 | — | 32 | ||
| @scaling01Nvidia released Nemotron 3 Super - a 120B-A12B hybrid Mamba model with LatentMoE and MTP -… | AI | 200 | 9 | — | 31 | ||
| @scaling01nice try, but it's just empty words show us hundreds of billions for infrastructure and da… | AI | 196 | 23 | — | 33 | ||
| @scaling01even more reason for frontier labs to just stop releasing small models not only are they g… | AI | 196 | 11 | — | 34 | ||
| @scaling01Anthropic, IPO and SpaceX are all racing for liquidity if one of these IPOs flops then the… | AI | 192 | 9 | — | 31 | ||
| @scaling01back in my days we would have called GPT-5.1 just GPT-5-2025-11-12 or GPT-5-latest but wha… | AI | 188 | 20 | — | 32 | ||
| @scaling01I've had this suspicion for a while makes me less bullish on moonshot, but honestly i don'… | AI | 176 | 17 | — | 31 | ||
| @scaling01Google introduces Gemini 2.5 Computer Use (now available as preview in the API) super stro… | Engineering | 163 | 6 | — | 29 | ||
| @scaling01"Lisan you are so biased, always trash talking OpenAI and Sam. you're just a hater." no, I… | AI | 160 | 9 | — | 30 | ||
| @scaling01I think GPT Images 1.5 is slighty slightly worse than Nano Banana Pro but that means good … | AI | 154 | 14 | — | 30 | ||
| @scaling01xAI reportedly spent $20B in 2025 and revenue only grew by about 22% that year (vs somethi… | AI | 153 | 7 | — | 29 |