Pinned Tweet
Introducing Manta, our long-context model.

Everything you give it, it keeps. Manta reads a hundred million tokens in one prompt.

Early access opens soon: empero.org/manta

🧵
92
173
1,650
469,327
Since everyone is looking for a new agent.. 👀
🚀 Abacus Desktop 0.1.0 is out.

The abacus coding agent, now in a window: your projects and sessions, the agent's work as it streams, the diff, a terminal, and a browser it can use.

Open source. macOS, Windows, Linux. Any model.

abacus.empero.org
3
4
20
66,293
Replying to @EmperoAI
It's also mobile x.com/EmperoAI/statu…
Your coding agent shouldn't stop when you leave your desk.

📱 abacus 0.6.5 puts your terminal sessions on your phone. Watch it work, approve a diff, answer its questions or steer the next step. Your machine stays in charge.

github.com/empero-org/aba…
0
0
3
168
Empero retweeted
Manta is coming.

It took a lot of work from a small team, and a lot of runs that didn't work before the ones that did. I'm proud of where it landed: a hundred million tokens in one prompt, and it still finds what you hid in there.

More to announce soon.
Introducing Manta, our long-context model.

Everything you give it, it keeps. Manta reads a hundred million tokens in one prompt.

Early access opens soon: empero.org/manta

🧵
5
3
40
2,483
Empero retweeted
Models should be built on the knowledge of humanity. They should not be biased towards anything by force. A model should form its opinion based on the facts it has. A model will never serve a country, a model will always serve humanity.
0
1
11
447
Replying to @EmperoAI
Many questioned how well Manta performs under actual load.

So we gave it some: a real 7.4-million-token coding session, logged from a current frontier agent. 31,450 messages of training runs, logs and fixes.

We cut it, and let Manta write the next turn. ↓
0
0
2
1,117
Many questioned how well Manta performs under actual load.

So we gave it some: a real 7.4-million-token coding session, logged from a current frontier agent. 31,450 messages of training runs, logs and fixes.

We cut it, and let Manta write the next turn. ↓
8
16
79
292,551
Replying to @EmperoAI
One sample per cut. Nothing edited, nothing picked.

Every turn sits next to the frontier agent’s, so you can judge for yourself. IPs, keys, paths and internal names are redacted; nothing else is changed.
1
0
6
292
Replying to @EmperoAI
Still legible at 7.4 million tokens.

Read every turn: empero-org.github.io/manta-agentic-…
JSON + Markdown: github.com/empero-org/man…

Manta is in research preview. Join the waitlist: empero.org/manta
0
0
6
286
Empero retweeted
Introducing Manta, our long-context model.

Everything you give it, it keeps. Manta reads a hundred million tokens in one prompt.

Early access opens soon: empero.org/manta

🧵
92
173
1,650
469,327
All Manta Reviews will happen during the next week. We have already received way more requests then anticipated, some organisations with the best fit might get a early reply. Please be patient and thank you all for your support. We will release agent continuation excerts today!
1
0
23
1,053
Replying to @EmperoAI
These are recall tests: they show that Manta finds a specific fact in a very long input. The rows, the logs and every answer are public:

github.com/empero-org/man…

More results come with the public release.
2
2
63
15,674
Replying to @EmperoAI
Next: a billion tokens.

We will open Manta to a small group first: researchers, engineers and teams with work that needs a model to keep a lot in mind at once. Tell us what you'd do with it:

empero.org/manta
4
0
100
12,118