I tink, therefore I am Post-training API by @thinkymachines
San Francisco
Joined January 2026
- Tweets183
- Following1
- Followers15.3K
- Likes254
Tinker retweeted
Tinker lets customers turn their expertise into cutting-edge models that they own. One year after launch, we look back at what we’ve shipped and what our users are building.
thinkingmachines.ai/news/one…
Tinker retweeted
Thinky is my go to for this thinkingmachines.ai/tinker/
Tinker retweeted
Today, we are launching Tinker grants of up to $50,000 in credits for safety research on open-weight models. We share some project ideas that excite us below; if you’re working on a safety project that could be accelerated by additional Tinker credits, we want to hear from you!
Tinker retweeted
Kimi K3 + Tinker is a really nice setup for an autoresearch loop
With most of the training plumbing off the table, Kimi’s code changes stay focused on the experiment itself. This makes it easier to follow what changed and steer the next runs.
Reproducing the Self-Distilled RLVR paper, Kimi one-shots the baseline implementation, ran 19 experiments across six configurations, and wrote the final report (plus a bonus, unsolicited Chinese version)
Tinker retweeted
Thinking Machines is impressive. In a couple hours I just fine tuned my own Qwen3.5-397B model this afternoon.
Fast usable multimodal is also going to enable very mind-blowing personal AI.
People talk, listen, watch, think, and collaborate at the same time, in real time. We've designed an AI that works with people the same way.
We share our approach, early results, and a quick look at our model in action.
thinkingmachines.ai/blog/int…
Tinker retweeted
People talk, listen, watch, think, and collaborate at the same time, in real time. We've designed an AI that works with people the same way.
We share our approach, early results, and a quick look at our model in action.
thinkingmachines.ai/blog/int…
Tinker retweeted
We built a new task to test AI research capabilities! Agents asked to use @tinkerapi from @thinkymachines to train a model on logic games. That involves writing full training pipeline, running experiments across recipes, and submitting the best model.
FrontierSWE was built with collaborators from industry and academia to ensure that tasks are diverse and reflect real work engineers and researchers encounter.
We specifically thank our partners @Modular, @PrimeIntellect and @thoughtfullab for their contributions
Tinker retweeted
Great work by Chroma training a search agent with SoTA efficiency. Lots of cool details: a prune tool for editing context mid-search, a synthetic data pipeline with verification steps, and a curriculum that shifts from recall to precision. Trained with Tinker!
Tinker retweeted
Introducing OpenReward.
🌍 330+ RL environments through one API
⚡ Autoscaled sandbox compute
🍒 4.5M+ unique RL tasks
🚂 Works like magic with Tinker, Miles, Slime
Link and thread below.
Tinker retweeted
Models that are great at calibrated predictions will be transformative for decision making. Excited about Mantic's work and proud they're using Tinker. Their new blog post digs into their methodology and findings.
I always dreamed of AGI as a wise advisor for humanity. Although LLMs are great for coding & knowledge work, I wouldn’t trust them to give me advice on my career, business strategy, or policy preferences. How can we build AI systems optimized for wisdom?
At Mantic we believe the unlock is prediction: predicting world events as accurately as possible, and hill-climbing this single metric.
Today we share some recent progress on the Thinking Machines website, having found Tinker a great platform for our RL experiments.
TL;DR: We RL-tune gpt-oss-120b to become a better forecaster than any other model. Having good scaffolding is a prerequisite. A fun result: our tuned model + Grok are decorrelated from the other best models, and so are the most indispensable when picking a team.
Tinker retweeted
There's a lot of sophisticated users out there with very interesting use cases who are more bottlenecked by infra (hard and annoying) than by data or RL science (fun!). At some point I participated on @ManifoldMarkets a ton (I was #6 in the first Masters league season :P manifold.markets/leagues/3/m…) so it's fun to see Tinker being used for forecasting/predictions
I always dreamed of AGI as a wise advisor for humanity. Although LLMs are great for coding & knowledge work, I wouldn’t trust them to give me advice on my career, business strategy, or policy preferences. How can we build AI systems optimized for wisdom?
At Mantic we believe the unlock is prediction: predicting world events as accurately as possible, and hill-climbing this single metric.
Today we share some recent progress on the Thinking Machines website, having found Tinker a great platform for our RL experiments.
TL;DR: We RL-tune gpt-oss-120b to become a better forecaster than any other model. Having good scaffolding is a prerequisite. A fun result: our tuned model + Grok are decorrelated from the other best models, and so are the most indispensable when picking a team.
Tinker retweeted
Tinker 🤝 Harbor
New in the cookbook: Harbor RL trains models on real software engineering tasks inside sandboxed containers. The agent gets a bash shell, an instruction, and a test suite. If the tests pass, it gets the reward.
github.com/thinking-machines…
Tinker retweeted
I joined @thinkymachines to work with @johnschulman2, @miramurati, @soumithchintala, @dchaplot, @alexgartrell, and others on the future of collaborative AI!
And I can confirm that we do have 45 lb weights.
If you're interested in training weights and/or Tinkering with us, join us! thinkingmachines.ai/#join-us
Tinker retweeted
🚨 New: Integrating Harbor (@harborframework) for end-to-end Computer-Use evaluation(for Windows and Linux) at scale with @thinkymachines' Tinker, OSWorld, @daytonaio, and bare-metal servers.
We just added support for Computer Use, @tinkerapi, and OSWorld to Harbor - a framework for evaluating agents and generating RL training data by running large-scale rollouts across parallel sandboxed environments and collecting trajectories for SFT and RL.
Repo and blogpost below 👇
Tinker retweeted
Qwen 3.5 is🔥, and now it's available on 🚀!
Four Qwen 3.5 models from @Alibaba_Qwen are now live on Tinker. Qwen 3.5 introduces hybrid linear attention that enables long context windows, as well as native vision input.
Tinker retweeted
Had fun building with @tinkerapi @thinkymachines
Multi-agent math coach — one Qwen3-8B as Solver, Coach, and Reviser, RL fine-tuned on competition math.
Sparse rewards, noisy gradients, fast policy drift. Self-correction is a harder optimization surface than solving. The model fixes formatting perfectly but algebraic and logical errors stay near zero.
Math RL hits different.
mohammedalshehri.blog/post/t…
Tinker retweeted
Why Tinker?
The @tinkerapi is simple (forward_backward, optim_step, sample, checkpointing) yet expressive to implement any weight-updating algorithm you want. It also hits the right boundary, pushing infrastructure logic below the API and algorithmic control fully above.
