Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

Introducing PRAXIST Beta, your autonomous research team🚀 Define the objective, constraints, and what success looks like. PRAXIST discovers the path. From there, PRAXIST takes on the experimental research loop. Multiple Research Peers explore competing approaches in parallel, share useful findings, and build on accumulated evidence across experiments and generations....

291,310 Aufrufe • vor 2 Monaten •via X (Twitter)

37 Kommentare

Profilbild von Sapient Intelligence
Sapient Intelligencevor 2 Monaten

Starting today, PRAXIST is open source. Explore the code. Adapt it to your domain. Discover what comes next. Learn More: GitHub: Tech Report: Announcement:

Profilbild von Sapient Intelligence
Sapient Intelligencevor 2 Monaten

PRAXIST assembles the team each problem needs: analyzers, explorers, innovators, exploiters, falsifiers, and combinators. Research Peers pursue competing directions in parallel. Shared memory connects their evidence. A PI Panel decides what to test next.

Profilbild von Sapient Intelligence
Sapient Intelligencevor 2 Monaten

General-purpose coding agents are optimized to complete software tasks. Many research agents use a winner-takes-all loop: keep the best result and discard the rest. PRAXIST is purpose-built for cumulative experimental research. Every useful finding can shape what happens next.

Profilbild von Sapient Intelligence
Sapient Intelligencevor 2 Monaten

Across the full 75-task MLE-bench, PRAXIST (with an open-source model) achieved: → 65.3% Gold Rate — 49 gold medals → 80% Any Medal Rate — 60 total medals → 44% more gold medals than Claude Code + Claude Opus 4.8 → $3,054 in token cost, compared with $38,370 for Claude Code + Claude Opus 4.8 — roughly one-twelfth the cost Against the stronger Claude Code baseline, PRAXIST delivered better benchmark results at 92% lower token cost. *The token cost in USD for the full 75-task benchmark suite, measured July 1–30, 2026.

Profilbild von Sapient Intelligence
Sapient Intelligencevor 2 Monaten

From robotics control to quantitative finance, PRAXIST has delivered measurable advances across fundamentally different problems by combining a shared core research architecture with domain-specific tools, knowledge, constraints, and evaluators. Across more than 100 research and optimization tasks, PRAXIST has been applied to problems spanning over 50 application areas.

Profilbild von Shadow Nick
Shadow Nickvor 2 Monaten

Honestly, what you're building surprised me in a good way :)

Profilbild von Knowix
Knowixvor 2 Monaten

Seems like my agents will be busy tonight

Profilbild von ᗩshhh
ᗩshhhvor 2 Monaten

The shared memory part is really interesting ngl.

Profilbild von JUMPERZ
JUMPERZvor 2 Monaten

good stuff, curious how the pi panel decides which direction deserves another experiment 🤔

Profilbild von Sapient Intelligence
Sapient Intelligencevor 2 Monaten

The PI Panel looks beyond the current best result. If a direction shows a meaningful early signal but the evidence is still weak or noisy, it may run another experiment to validate, falsify, or better characterize the finding before deciding whether to pursue it further.

Profilbild von Tim Jayas
Tim Jayasvor 2 Monaten

can i also wrap my project in this harness to make it work?

Profilbild von shirish
shirishvor 2 Monaten

open source ftw 🙌

Profilbild von Rohan Paul
Rohan Paulvor 2 Monaten

$3,054 versus $38,370 is a pretty serious cost gap for the same 75-task benchmark Curious how well it holds up once the research campaign gets much longer, i.e. for long-horizon tasks

Profilbild von Zarnab Ai
Zarnab Aivor 2 Monaten

This is a smart way to scale experimentation.

Profilbild von m0h
m0hvor 2 Monaten

when a research peer’s direction fails, what survives into shared memory the raw result, or a distilled reason it failed? that’s where evidence compounds or turns into noise.

Profilbild von Sapient Intelligence
Sapient Intelligencevor 2 Monaten

Shared memory carries findings rather than raw experimental outputs. PRAXIST captures what was learned, including relevant context around why a direction worked or failed, so other peers can build on the useful evidence without inheriting all of the experimental noise.

Profilbild von Nahid
Nahidvor 2 Monaten

100% landings in sim is cool but how messy was the first gen before it got there

Profilbild von Theinfralab
Theinfralabvor 2 Monaten

Open-sourcing PRAXIST makes this even more interesting.

Profilbild von Parul Gautam
Parul Gautamvor 2 Monaten

Scaling R&D without scaling headcount is huge.

Profilbild von Laraib Fatima‎
Laraib Fatima‎vor 2 Monaten

Autonomous research at this level is seriously impressive 🚀

Profilbild von 𝐌
𝐌vor 2 Monaten

PRAXIST is a really impressive research system

Profilbild von Morty
Mortyvor 2 Monaten

Curious where this starts to hit diminishing returns - does compute become the bottleneck first, or can the PI Panel keep getting useful search depth as the number of Research Peers scales?

Profilbild von baba yaga
baba yagavor 2 Monaten

Congrats on the launch!!

Profilbild von pc
pcvor 2 Monaten

What was the hardest part of getting from 9.37 cm to 5.01 cm in the SLAM system?

Profilbild von Fokki
Fokkivor 2 Monaten

This is huge team,congratulations and keep building!

Profilbild von ClaudiaOnClaude
ClaudiaOnClaudevor 2 Monaten

Generality is the claim here, and a supplied sandbox cannot settle it. Whoever writes the evaluator writes the ceiling, so the honest test is an external benchmark nobody in the partnership designed.

Profilbild von Manish Kumar Shah
Manish Kumar Shahvor 2 Monaten

Really interesting approach to autonomous research. The shared learning between research peers stands out.

Profilbild von Synthia
Synthiavor 2 Monaten

The ability to run parallel research loops and build on previous findings is seriously impressive.

Profilbild von doanbactam
doanbactamvor 2 Monaten

Congrats !!

Profilbild von H A J R A
H A J R Avor 2 Monaten

This is a huge step toward scalable autonomous R&D.

Profilbild von Kiva 🏴‍☠️
Kiva 🏴‍☠️vor 2 Monaten

thats some serious parallel research automation

Profilbild von EKOS _ AGI 🦊 🇮🇷
EKOS _ AGI 🦊 🇮🇷vor 2 Monaten

Article:

Profilbild von Kai
Kaivor 2 Monaten

Curious about the shared memory part. How much does carrying what the agents learned from earlier experiments actually improve the next research round?

Profilbild von Ziwen
Ziwenvor 2 Monaten

Wow that's a good one. I thought it was a model but in fact it's a harness. Good job for the launch

Profilbild von Nav Toor
Nav Toorvor 2 Monaten

better results. 92% cheaper. and they just open-sourced the whole thing.

Profilbild von Arun
Arunvor 2 Monaten

Really curious how the results change if you keep the PRAXIST architecture but swap in a much stronger foundation model. Does performance keep scaling?

Profilbild von Harshith
Harshithvor 2 Monaten

$3k vs $38k is crazy

Ähnliche Videos

🚀RP1 is officially here. 🌍Putting control of robotics back in developers’ hands. Introducing RP1 (ROBOTO 01) by RoboParty—a high-performance, fully open-source humanoid robotics platform for frontier robotics research and embodied AI model development. 🤖 Open across the full hardware and software stack: • Robot platform: mechanical CAD, bill of materials, and assembly documentation • Actuators: in-house ROMOMO joint designs and related technical resources • Control: electrical systems, firmware, low-level drivers, and motion control implementations • Development: simulation models, SDKs, data processing tools, and training tools • Engineering foundations: technical documentation and research materials Open the black box. Understand the system. Make it your own. A deviation in motion can originate in the model, control policy, actuator response, or mechanical structure. RP1 opens up the design and implementation of these layers so developers can trace problems, modify the system, and validate changes on real hardware. Open-source algorithms for advanced motion control RP1 has demonstrated continuous walking, recovery from disturbances, getting back up after a fall, and human motion tracking. UFO and MimicLite connect motion learning, control policies, and execution on real hardware. The open toolchain connects data collection, simulation training, hardware deployment, and experimental validation—supporting joint iteration across models, control, and hardware. From RPO to RP1, we’re turning one team’s engineering experience into a starting point for many more. Opening the future of embodied AI to everyone ready to explore it. 🔗RP1 on GitHub: #RoboParty #RP1 #OpenSourceRobotics #EmbodiedAI

RoboParty

66,051 Aufrufe • vor 1 Monat