dan ./ retweeted
DeepSeek V4.1 Flash (1017) solved 8/17 Zero2Repo tasks with dsh, at $3.18 in reported API spend across the full suite.
OpenCode scored 7/17 and Claude Code, 5/17.
Costs are reported API spend across all 17 tasks.
Check out more findings: zero2repo.ai/
dan ./ retweeted
SWE-bench Verified has become saturated. Seven models scored 95–97% (Vals AI, Sep 2026). But how well can coding agents build complete software projects from scratch?
Zero2Repo is a Gradient Data research initiative building hard, verifiable tasks and training data for whole-project development.
Agents start with a full spec, an interface contract and an empty workspace. They build the entire repo without reference code, then submit it for hidden deterministic testing. Every test must pass for a task to count. No LLM judge is used.
The first release includes 17 tasks derived from real open-source projects across Python, TypeScript, Go, C, C++ and Zig.
Across 18 model–harness configurations, scores range from 1/17 to 15/17. The top 10 span 47 percentage points, compared with 5 points on SWE-bench Verified.
No configuration has completed all 17 tasks. This matters for self-iteration: small unresolved issues can persist and compound as agents continue developing a project. Reliable self-evolution remains an open research challenge.
We’re expanding the task collection and looking for contributors to suggest seed repositories, develop tasks and review requirements and hidden tests.
More cases are coming.
👉 zero2repo.ai
👉 github.com/Gradient-Data/Zer…
dan ./ retweeted
One area we’re focusing on is software engineering.
Seven models scored 95–97% on SWE-bench Verified. (Vals AI, Sep 2026) But can they build full projects from scratch?
Zero2Repo, Gradient Data’s latest research initiative, builds verifiable tasks for training and evaluation.
zero2repo.ai
dan ./ retweeted
Over the past year, we’ve worked on Echo and Logits to make AI post-training more efficient and accessible.
Gradient Data now extends that work with verifiable environments, real-world trajectories and evaluation systems for more capable and reliable agents.
gradientdata.ai
dan ./ retweeted
A few research milestones worth looking back on as September comes to a close.
Our team and collaborators have been working across model infrastructure, agents, and evaluation, from making RL more efficient to testing whether AI systems actually do what they intend.
🧵
dan ./ retweeted
Agents are impressive. Multi-agent systems are where things get interesting.
As AI systems evolve from single agents into networks of specialists, a new question follows:
How do agents decide who does what?
That’s a problem we’ve been exploring with Symphony.
Symphony brought multi-agent collaboration across heterogeneous environments. To take that further, we released Symphony-Coord, where coordination and specialization adapt through task context, performance, and feedback.
github.com/GradientHQ/sympho…
dan ./ retweeted
Can web agents learn through imagination?
DynaWeb: Model-Based Reinforcement Learning of Web Agents has been accepted to EMNLP 2026 Main Conference.
Training web agents with online RL means costly, risky interaction with the live web. DynaWeb learns a Web World Model instead, letting agents dream web interactions and train on imagined rollouts.
- Learns naturalistic web transitions from agent actions
- Generates imagined trajectories in a synthetic web environment
- Interleaves imagined rollouts with real expert trajectories
- Consistent gains for open-source agents on WebArena and WebVoyager
Imagination as a scalable path toward online agentic RL.
This is part of our ongoing work at Gradient on distributed and efficient RL: training capable agents without the cost, latency, and risk of centralizing them.
Paper: arxiv.org/abs/2601.22149
Code: github.com/jadeleiyu/MBRL-Ag…
dan ./ retweeted
Proud to share MERA: joint work from a group of brilliant researchers including our very own.
It evolves small models with SkillBook + routing + adapters so they take on more of the agent workload over time.
Paper + code 👇
🚀 MERA: Model Evolution and Routing with Skill Adaptation for Agentic Systems at Scale
How can we make small models stronger for on-device agent deployment?
MERA uses stronger models to guide an iterative loop of RL/GRPO, skill learning, and router optimization. Student failures become verified demonstrations, reusable SkillBook procedures, and LoRA updates, helping the small model take on more work over time.
🔥 Results:
Qwen2.5-Coder-1.5B: 28.7% → 49.7% coding pass
Qwen3.5-2B on TAU-2: 14/35 → 18/35
Fine-tuned 2B matches an unadapted 4B model
Don’t just route around small models. Evolve them.
📄 arxiv.org/abs/2608.10333
💻 github.com/yh-yao/MERA-Evolv…
dan ./ retweeted
🧵 𝗪𝗵𝗮𝘁 𝗰𝗮𝗻 𝘆𝗼𝘂𝗿 𝗔𝗜 𝗔𝗴𝗲𝗻𝘁 𝗮𝗰𝘁𝘂𝗮𝗹𝗹𝘆 𝗱𝗼? 𝗟𝗲𝘁’𝘀 𝗯𝗿𝗲𝗮𝗸 𝗶𝘁 𝗱𝗼𝘄𝗻:
1/ The Teneo AI Agent Testnet is now live. You’ve heard about “agents”.
Here’s how they actually work 👇
dan ./ retweeted
Bless <> Warden
Bless will power @wardenprotocol AI agents
💻 Warden enables verifiable AI across chains
🌅 We provide the scalable, low-latency compute to make it possible.
Together, we’re reclaiming the AI stack.
dan ./ retweeted
We recently ran the first-ever Bless Network Film Festival.
The Blessheads showed up with incredible submissions.
But two stood above the rest.
Here’s the first of our winners: @perpetual1ty and his film “Sacred Relic”
Staying up stitching sounds to scenes, crying over Blender crashes. Doubting, redoing, deleting — and then somehow, it all paid off.
I am grateful to @theblessnetwork for the opportunity and providing the stage to show what I could do. I learned so much & I can’t wait for more.
dan ./ retweeted
Join our community pope @thomlovesyou tomorrow as he yaps it up with Kaito’s FS7 in the 0.1 Yaps Club.
We’re talking all things Bless and the people powering it. 🌅
dan ./ retweeted
𝗧𝗵𝗲 𝗔𝗜 𝗔𝗴𝗲𝗻𝘁 𝗧𝗲𝘀𝘁𝗻𝗲𝘁 𝗶𝘀 𝗟𝗜𝗩𝗘 🚨
Today marks a huge milestone for Teneo: the AI Agent Testnet is officially live.
Access is invite-only, and the journey begins now.
Want in? You will need a Golden Ticket.
They are in circulation — passed from early testers to new ones.
🧠 Redeem the ticket with a linked wallet + Machine ID
🔓 Process 10,000+ content items → unlock 3 new invites
This isn’t open access. It’s earned access.
Host your Agent. Then empower the next wave. AI Starts With You.
🔗 Read all details: teneo.pro/blog/activating-th…
If your device overheating or what, it is nothing to do with @theblessnetwork and the important thing is our device really matter, we used it everyday, it's our responsibility to take care of our devices.
Computer overheating? 🫠
It’s not your Bless extension!
We assess device capacity in real time and borrow a small percentage of the compute resources that aren’t being actively used.
4
2
59
















