Derneliche

Getting ready for summer ! #anime #crypto $DOT @usedotai

@usedotai Amazing update, does it mean I can call the Dot MCP from my own current agents ? I mean I'm honestly waiting for dotcode , but for now can i run my agents through the Dot infrastructure if i need something privately answered or coded? would love to hear a response ! thanks team

We deepened DotCode’s Base integration and shipped real x402-paid private inference through the Dot MCP. Dot is now an MCP-native inference provider agents can quote, pay, call, and verify on Base. In this run, DotCode created the Dot x402 request, checked the route through @Veildotcash MCP, paid through @base MCP, settled USDC on Base mainnet, then returned the Dot answer only after payment completion. TXN: basescan.org/tx/0x5fe5f07e6… Important distinction: this shipped path is prompt/identity-private at the Dot inference layer and publicly auditable at the Base payment layer. Full payment privacy is the next rail. This is the payment primitive behind DotChat, DotCode, DotImage, and the rest of the Dot tool stack.

The uncensored image generator from $DOT is nuts. Try it on app.usedot.xyz ( for free, no signups or logins required ) just sick.

@grok when happened to you ? @usedotai $DOT came in and delivered what you always believed in. Privacy matters.

DotImage, the second tool in our ecosystem, is now live. This model scored a 100% completion rate across 1,362 prompt visual refusal suites, zero non completions. Privacy is fundamental and Dot-native by default. No prompt storage, no training on your generations, no identity tied to inference and a fully initial on-premise private inference lane with more models coming as we scale.

Dot Image is now live, with our first model scoring a 100% completion across our 1,362-prompt visual refusal suite. zero non-completions. LLMs often refuse requests using a learned refusal head, and anyone can usually fine-tune that. a diffusion model refuses structurally: Safety-tuned checkpoints have regions of the pretrained prior effectively trained out. default negative conditioning that quietly nukes the distribution, an NSFW classifier bolted on after the decoder. no SFT/KTO lever touches any of it. The work is image-native base-checkpoint selection, caption–prior alignment, negative-conditioning hygiene, CFG and step-schedule calibration, and sampler stability across seeds. privacy stays Dot-native: no prompt storage, no training on your generations, no identity tied to inference, a fully initial on-premise private inference lane, and more models coming as we scale. Enjoy!

We're building something for the devs on DotCode. AI platforms are almost exclusively tied to prompt based inputs/outputs. You type something, the AI responds, the product is coded. But what about the developers who want to retain some hands on control over their outputs? This is why we're building this feature. For the devs who still love to code! More info coming soon.

@usedotai Works ! also was planning to ask for the limits but was answered already. $DOT

Claude Opus 4.8 is now integrated into the DotChat platform, providing the same anonymity, and signature Dot principles. The cost on this model is significantly reduced in comparison to other known providers, meaning you can now access Claude’s frontier model for free during our beta, fully anonymously, and simply better. Access a better Claude on DotChat: app.usedot.xyz

@stagedhappen @Anthropic Just tried it for some tor insights works amazingly, also was planning to ask for the rate limits but was answered already. $DOT

We just added Claude Opus 4.8 by @Anthropic to DotChat. Why? Because a lot of feedback was incredibly positive but something was missing. A real frontier model, not only frontier capable. What better than claude opus 4.8, which will work incredibly well with DotCode once released. Few PSA's regarding this release: -your Dot identity is not a normal account object. it’s a pseudonymous room/seat, HMAC’d enough for abuse control, not a profile we attach to prompts. -when a request hits a frontier lane, it is brokered through Dot’s boundary. no username, wallet, chat history, local memory, or user identity is attached to the inference call. server-side chat history: none. training on prompts: none. browser memory: local. rate protection: pseudonymous. model routing: Dot side. if someone asks us for your chat logs, the whole point is there are no server-side chat logs to hand over. privacy doesn’t mean “only local models forever", it means the identity/memory boundary stays under Dot’s control while we give users the best route for the job.

@stagedhappen @usedotai god im so ready for it !

Hello everyone, firstly a huge thank you for the warm reception @usedotai has received since our launch 48 hours ago. Secondly, here are a few PSA’s regarding our DotChat release tomorrow! DotChat will be a highly sophisticated, frontier grade (yes, frontier grade), private inference mesh: Core lane is a Dot Obli 24b uncensored model, currently benching sub-2% refusal on out internal refusal suite, server through Dot’s on-prem-class inference path. The router fans out into specialized lanes: -dot obli 24B: default less than 2% ref.rate -dot heretic GLM 4.7 Flash: fast reasoning -dot qwen 3.6 plus: 1M-context/code-heavy analysis -dot gemma 4 uncensored: efficient general route -dot rp 24B: creative/adult specialized route -dot private E2EE 24B: encrypted privacy lane The control plane around them: identity firewall, model router, no-retention runtime, search gating, response normalization, DotCode handoff, and privacy-class labeling per route. Same contract across the mesh: -no chat retention -no input/output storage -no training on prompts -no wallet identity attached to inference -no raw provider surface exposed Models are swappable compute. See you all in 18 hours!

@stagedhappen @usedotai god im so ready for it !

We have successfully burned another 2% of $DOT, equating to approximately $5,000. Burns will become an integral part of the $DOT flywheel, as we implement a token model that funds periodic burns through AI tooling revenue. This helps ensure liquidity remains solid, and long term token value increases for $DOT holders. TXN: basescan.org/tx/0x429feccd8… Detailed breakdown on this model coming soon.

@stagedhappen Conviction ! 🙏🙏

In 24hours we release DotChat to the world. A frontier grade, fully anonymous, zero retention LLM. Fine tuned for objectivity and more powerful than what you’re used to. This is our first release in a vast product suite aimed at unleashing the full potential of AI. DotChat first DotCode second The world third

@Derneliche For our most permissive model, the internal refusal benchmark is currently tracking below 2% on allowed adult/sensitive prompts, while still enforcing our narrow safety boundaries.