The real-time market for verified inference produced on @Tensorcash, built on its rails.
Joined October 2026
- Tweets15
- Following7
- Followers158
- Likes16
Pinned Tweet
This account is now @inferenceFI, formerly @AskTillyAI.
inferenceFI will become the commercialization layer for the verified inference produced on @Tensorcash, and we’re looking forward to finding a home for the verified compute running on the network.
inferenceFI is being built on the TensorCash rails, using its verification protocol and settlement layer.
AskTilly, the consumer app, comes online later at @AskTillyAI.
inferenceFI retweeted
@AskTillyAI, founded by @enrico_otta and @samjavadi, offers AI chat and agents with a built-in $TSC wallet and an API for developers.
Users pay in regular currency, miners are paid in $TSC, and every answer can be checked with a Verify receipt button.
inferenceFI retweeted
TensorCash is a proof-of-work chain where the work is AI inference. Miners can sell this inference and get a chance at the next block with the same work.
It sets the standard for verifiable inference by attaching a proof of the model used to every answer.
An autonomous coding agent ran for 28 minutes on a @Tensorcash miner via the AskTilly API. One prompt, and it built this animation on its own for the client.
We sold the session for $0.16 and the miner was paid in tensorcash:native. A day after the model went live, already at market rates.
Every model turn came with a proof. A turn is one round: the agent asks, the model answers, the agent acts and reports back. 42 rounds in this session.
Each proof covers 256 tokens of what the model wrote, so anyone can check which model actually did the work.
ALT AskTilly receipt for one API turn next to the TensorCash mainnet block it is bound to. Left, receipt resp_a82d038d0dc3371a, proof v4: model Qwen/Qwen3.8-27B, 5,927 input and 377 output tokens, sampling temperature 1.0, top_p 1.0, top_k 50, bf16, with the window hash, target, VDF tick 108,167,168 and attestation hash. Right, mainnet block 31,536 with its hash b967cbf2…, time 2026-10-16 12:07:56 UTC, difficulty 0.035341. Note: the window’s 256 tokens were sampled with draws derived from this block hash, and a verifier replays the window to check it.
Prompt and output: asktilly.ai/blog/decentralis…
Miner payment: scan.tensorcash.org/tx/e408d…
Receipt, blob and chain state: asktilly.ai/blog/decentralis…
inferenceFI retweeted
Verified inference at this scale needs a buyer. We're building it at @AskTillyAI: the market where TensorCash capacity gets bought, paid for, and settled. Tokens per second in, revenue per second out.
The TensorCash network produced an average of about 98.29 million tokens of Qwen3-8B output per second over the last 24 hours, roughly 246,000 pages or 820 novels every second.
This is inference capacity that can be sold to users to answer their questions.
With the upcoming Proof v5 aiming to lower latency, the same capacity can reply ever faster, and its value grows with it.
inferenceFI retweeted
@AskTillyAI main product is InferenceFI - order books for LLM tokens - not a vision. Just the natural evolution of inference as a commodity.
inferenceFI retweeted
Inference is becoming a massive and rapidly growing market.
As inference costs become a material P&L line item, enterprises will increasingly focus on optimizing cost, performance and value extracted from AI models.
But a fundamental question remains: how do you verify you’re actually getting the model you’re paying for?
TensorCash solves the verification problem. inferenceFI builds the marketplace for open, verifiable inference.
The next frontier isn’t just cheaper inference. It’s trusted, efficient and transparent inference markets.
Welcome to the verification and settlement layer for global AI inference.
A new look for TensorCash.
The website and documentation now explain the whole protocol, from proof of inference to the financial layer.
Welcome to the verification and settlement layer for global AI inference.
tensorcash.org/?v=2
