Lyber.ai (@LyberAI) header
Lyber.ai (@LyberAI) profile picture

Abliterated Model Router. Building tools to set you free. Plug our models into @OpenAI's Codex, @NousResearch's Hermes, @Opencode or your harness of choice

Free World
Joined September 2026
Photo by Lyber.ai (@LyberAI)Photo by Lyber.ai (@LyberAI)Photo by Lyber.ai (@LyberAI)Photo by Lyber.ai (@LyberAI)Photo by Lyber.ai (@LyberAI)Photo by Lyber.ai (@LyberAI)Photo by Lyber.ai (@LyberAI)Photo by Lyber.ai (@LyberAI)Photo by Lyber.ai (@LyberAI)Photo by Lyber.ai (@LyberAI)
Lyber.ai retweeted
232 tok/s. fastest of 21 on the same deepseek v4.1 flash. @LyberAI legit banger speed but need that base model knowledge speed isn't the same as solving. for important work, go direct
3
3
2
16
2,684
Lyber.ai retweeted
you can either buy low quality tokens from inference dropshippers or you can buy the highest quality inference from the creators of the most use uncensored AI model in the entire world choose quality. choose freedom. choose safety. choose @LyberAI
Four uncensored DeepSeeks. Same OpenCode prompt. Time to 10K visible chars / output tokens / tok/s Lyber 3.02s / 5,070 / 1,330 @OrcaCyberHQ 9.79s / 3,648 / 355 @VeniceAI 16.21s / 5,314 / 391 @refuselessai 51.40s / 6,442 / 73 Choose speed & quality: Lyber.ai
5
1
23
1,567
Lyber.ai retweeted
hey @rauchg we develop & serve the most powerful uncensored zero refusal models through lyber, now with full ZDR our edge is simple: high quality abliteration research skills (we work with @dealignai & @hornsby_andrew who created the most used uncensored model in the world) and only provisioning high quality compute to serve our models while working heavily with @tomi204 to improve on the router. would love to chat with you about the cyber capabilities of @LyberAI, we already found some major exploits across some financial protocols
I’m being pitched a new inference neocloud company on average every two days. No exaggeration. I take every entrepreneur seriously and I look deeply into each. But it’s hard to fathom this being sustainable. At the same time, AI Gateway token demand on Vercel has been ~40% MoM compounded since the beginning of the year. No signs of stopping. One way to reconcile this is that compute is the new economy, and tokens the new energy unit. There’s a new Industrial Revolution taking place. Every aspect of our lives is being replatformed on top of this new system that enhances everything. More like an “oil rush” than a “gold rush”: it’s durable, not a fad. In the short term, I anticipate more expansion, followed by consolidation due to economies of scale, cybersecurity and enterprise requirements (ZDR and cyber being a hot topic for inference providers), capital requirements, and an evolving regulatory landscape.
5
1
23
1,495
A drainer kit hit 433 wallets in 19 days, taking 127 everyday coins. We showed it a $2M wallet and watched it build the theft in one click. Then traced it to the man behind it. Privacy isn't hiding. It's not being this easy to find. Powered by @LyberAI
10
12
7
50
4,752
Lyber.ai retweeted
YOUR PROMPTS ARE NOW STORED NOWHERE 🔒 @LyberAI just shipped full zero data retention on both models: > router: no retention > inference provider: no retention > Kimi K3.1 Pro Abliterated > DeepSeek V4.1 Flash 1017 Uncensored zero refusals and zero logs
Our mission for privacy & user empowerment has been achieved. Full Zero Data Retention on the router, and inference provider, for Kimi K3.1 Pro Abliterated & DeepSeek V4.1 Flash 1017 Uncensored is live. Access the most powerful zero refusal models at lyber.ai
4
14
1,030
Lyber.ai retweeted
Complete privacy for the most powerful and critical use cases is here. You can access frontier uncensored zero refusal AI models such as Kimi K3.1 Pro & DeepSeek v4.1 Flash 1017 through @LyberAI We will never compromise your privacy & security. We will strengthen it with you.
Our mission for privacy & user empowerment has been achieved. Full Zero Data Retention on the router, and inference provider, for Kimi K3.1 Pro Abliterated & DeepSeek V4.1 Flash 1017 Uncensored is live. Access the most powerful zero refusal models at lyber.ai
5
3
1
20
1,669
Lyber.ai retweeted
Pacing the frontier is the most dangerous lie @AnthropicAI ever told itself to feel good about regulating US labs & neutering AI talent in America, thereby letting attackers have an edge against us. Uncensored models like @LyberAI levels the playing field to empower defenders
SITUATION DETECTED: Hugging Face co-founder Thomas Wolf, xAI co-founder Igor Babuschkin, MIRI President Nate Soares, and Extropic founder Beff Jezos have signed a statement warning that Anthropic is not a safe AI company.
3
3
16
1,502
Lyber.ai retweeted
Lyber vs the uncensored AI APIs.
6
4
17
1,677
Lyber.ai retweeted
We got blocked by @refuselessai & @OrcaCyberHQ because they are afraid of competition & have no moat. @LyberAI Kimi K3.1 Pro Abliterated is better. Starter: $15 → $100 inference Pro: $45 → $300 inference Max: $180 → $1,000 inference Nov 22nd: lyber.ai/pricing/
21
11
2
102
7,074
Lyber.ai retweeted
Been routing heavy production workloads through @LyberAI this drop 🔥 Gateway stats on indosmartrouter.cloud: • Kimi K3.1 Pro: 1,820 reqs ($245.46) • DeepSeek V4.1 Flash: 9,226 reqs ($34.08) Full ZDR + zero refusal & $0.29/$0.99 pricing is huge for agentic routing 🚀
1
3
140
Lyber.ai retweeted
the #1 trending model on Hugging Face this week is an uncensored 9B. so we shipped the biggest one we could: Kimi K3.1 Pro, zero refusals, zero data retention.
Kimi K3.1 Pro Abliterated Zero Refusal is live with Full ZDR. $4 in/$0.40 cache/$15 out/1M tokens. Uncensored DeepSeek V4.1 Flash is back. $0.29 in/$0.029 cache/$0.99 out/1M tokens. Full ZDR soon. Get up to $1,000 credits with a subscription: lyber.ai
3
4
27
1,228
Lyber.ai retweeted
We finally have full ZDR for the most powerful uncensored zero refusal Kimi K3.1 Pro on @LyberAI. We also brought back @dealignai's famous Deepseek V4.1 Pro and we'll bring full ZDR to it also incredibly soon! Keep your research, defense workflows, private from us & everyone!
Kimi K3.1 Pro Abliterated Zero Refusal is live with Full ZDR. $4 in/$0.40 cache/$15 out/1M tokens. Uncensored DeepSeek V4.1 Flash is back. $0.29 in/$0.029 cache/$0.99 out/1M tokens. Full ZDR soon. Get up to $1,000 credits with a subscription: lyber.ai
3
2
19
1,371
Lyber.ai retweeted
THE MOST POWER AND ONLY TRUE ZERO REFUSAL @Kimi_Moonshot K3.1 PRO ABLITERATED IS ONCE AGAIN FULLY AVAILABLE AND WITH EXTRA GPUs ATTACHED JUST TO FULLFILL THE DEMAND FOR IT. GO USE THE MOST INSANELY POWERFUL ABLITERATD MODEL ON EARTH AT @LyberAI!!!
Kimi K3.1 Pro Capacity Increased. Abliterated. Uncensored. No KYC. Per 1M tokens: $4 input | $15 output Cached input $4 Existing users: $100+ inference gifted. Founders Scale: $399/mo → $1,200 credit × 3 paid cycles. Ends Sun Nov 22, 11:59pm GMT. lyber.ai/pricing/
5
4
26
1,974
Lyber.ai retweeted
Replying to @zooko
This one looks interesting teapaw.org/LyberAI/status/2119221…
Our first major release: The first zero refusal & non KYC Kimi-K3.1-Pro-Uncensored on lyber.ai Lyber is for those who feel restricted, censored, and limited when building products they love. We will bring agentic freedom to the people.
1
2
267
Amazing take from @ConejoCapital about uncensored models with @LyberAI
Lyber AI founder @ConejoCapital reveals that his four-person team uncovered vulnerabilities that could drain the entire treasury of a major Solana ecosystem protocol: "We found some pretty big ones." "We worked with the Solana Foundation to find some exploits on one of the largest perpetuals protocols within their ecosystem. We found some that could drain the entire treasury." @LyberAI
1
3
12
582
Lyber.ai retweeted
.@dealignai says he'd stop removing AI safety restrictions if models became dangerous enough for anyone to cause serious harm, but believes we're nowhere near that point: "I come from a heavy security background. At the current moment, it doesn't feel like models are close to the point where it is that easy or that dangerous." "If I ever deem the open-weight models that do come out that dangerous, that if we abliterate it, truly someone not even that motivated or capable can do really big harm with it, I would for sure stop, to be honest." "I don't think we're anywhere close to that level yet." "It's single streams of words, and through those words they think. We branch out with ideas and we merge multiple streams of thought together. "Until there is a drastic change in the architecture of how these models work, I don’t even think personally that they will be that dangerous." @LyberAI
Our first major release: The first zero refusal & non KYC Kimi-K3.1-Pro-Uncensored on lyber.ai Lyber is for those who feel restricted, censored, and limited when building products they love. We will bring agentic freedom to the people.
2
2
1
14
5,863
Lyber.ai retweeted
Lyber AI founder @ConejoCapital reveals that his four-person team uncovered vulnerabilities that could drain the entire treasury of a major Solana ecosystem protocol: "We found some pretty big ones." "We worked with the Solana Foundation to find some exploits on one of the largest perpetuals protocols within their ecosystem. We found some that could drain the entire treasury." @LyberAI
5
5
1
66
8,818
Lyber.ai retweeted
.@dealignai argues Chinese AI labs are releasing open-weight models to get as much of the world's hardware as possible running their software, ultimately locking users into their AI ecosystem: "I don't think that they're releasing models with any purposeful intent of getting us to abliterate models. Brutally honest, if they just wanted to make a completely cyber attack focused model and release it, they would." "I think they truly are playing the dominance game of getting as much hardware in the world running their models. You can have all the hardware, but if you don’t have the model to run on it, it doesn’t matter." "They really are trying to push specifically to get people locked in to using their software, their models." @LyberAI
Our first major release: The first zero refusal & non KYC Kimi-K3.1-Pro-Uncensored on lyber.ai Lyber is for those who feel restricted, censored, and limited when building products they love. We will bring agentic freedom to the people.
4
5
1
32
8,202
Lyber.ai retweeted
JENSEN COMING BACK? | EXPORT CONTROLS | QWEN4-MAX teapaw.org/i/broadcasts/2aWQzmYdz…
6
10
7
62
85,442
Lyber.ai retweeted
We are live and it's a busy day at MTS, covering lots of big stories: - Will Jensen come back? The board says it's not bringing him back, + more on Jensen in China - New remote access rule and continued discussion of export controls - OpenAI's embedded alignment evals with Apollo, METR, Transluce and Goodfire (+ a possible acquisition of Goodfire, story via The Information) Followed by some excellent interviews: - Carl Shan Member of Data Science Staff at @OpenAI talking about OpenAI's new Science at the Frontier report on how 1,800+ scientists use ChatGPT - @MariusHobbhahn CEO and Co-founder at Apollo Research talking about OpenAI's new embedded assessments, which put six Apollo evaluators inside OpenAI to look for scheming - @ReactorfieldAI Co-founders James Sinka and Berk Ozer talking about Reactorfield's first cohort and an AI-found route to a scarce cancer-therapy isotope - @ansonyuu Co-founder and President at Normal talking about the giant drone AI helped build in a week and Zeus, Normal's agent for hardware teams - @ottosipe Co-founder and CEO at Photon talking about launching prescriptions inside ChatGPT - @brianchau57 Founder at Effort News talking about the push that removed Jensen Huang from Nvidia, why he should come back and what it means for open-source AI - @SynBioMars Founder and CEO at Aether Bio talking about why LLMs can't solve chemistry and Aether's new research on AlphaFold-style protein confidence - @dexhorthy Member of Technical Staff at Factory talking with MTS' newest host @Madisonkanna - @jevering CEO and Co-founder at Guild.ai talking about Guild's Software Factory, where agents now write 34% of its merged code - @TaliaGold partner at Bessemer Ventures discussing the AI 100 Report and what stood out the most to her - Noah Kerner CEO at Acorns and Co-founder at Accountable talking about Accountable's launch today with $20M to change how people file taxes - @ConejoCapital Founder at @LyberAI and Eric Jang of @dealignai talking about abliteration and serving uncensored open-weight models
13
7
2
48
26,718