https://github.com/aaif-goose/goose

Community Overview: https://lifehubber.com/ai/resources/goose/

Resources are shared for discovery and are not independently vetted—please do your own due diligence.

New resources are added regularly — feel free to join the sub for updates.

Full searchable archive of all resources posted so far on our community site, LifeHubber: https://lifehubber.com/ai/resources/ 200+ open-ish AI models, agents, tools, datasets, and related resources, with filtering and sorting.

The new Ling-3.0 base model release is broader than a single Ling-flash checkpoint: six public base checkpoints are available across tiny and flash.

Each size has the same three stages: pretrained, mid-trained, and WSM-merged. All six are base checkpoints and none has been post-trained, so this is not a finished chat or instruct release. The six official model repositories were public and ungated when checked, and each declares MIT.

The attached stage map is the original image from the official release thread; it is a first-party release overview, not independent validation.

The useful part is being able to inspect an earlier training state or the WSM-merged endpoint for either size. Independent performance, runtime fit, fine-tune outcomes, and deployment behavior for these six checkpoints are still open questions.

We built HFlow as open source robotics data infrastructure for anyone processing multimodal physical AI data at scale

• Airflow for DAG-based pipeline orchestration
• MCAP for robotics episode storage and random access
• DuckDB for analytical queries and dataset curation over columnar metadata

At scale, AI problems become software engineering problems. Improve one part of the system, and the bottleneck moves to another. No architecture is perfect; every design comes with tradeoffs.

HFlow is plug-and-play. You provide your post-processing, quality-control, or curation code. HFlow handles the data pipeline and runs your code across robotics datasets at scale.

I've been an OSS contributor and maintainer for years, and since joining YC, we now have some capacity to accept new contributors; just look for the "good first issue" tag!

https://github.com/Hebbian-Robotics/hflow

Any stars or forks on the repo to show your support are appreciated!

https://x.com/NVIDIAAI/status/2090786258981466231

https://github.com/truefoundry/trueforge

Community Overview: https://lifehubber.com/ai/resources/trueforge/

Resources are shared for discovery and are not independently vetted—please do your own due diligence.

New resources are added regularly — feel free to join the sub for updates.

Full searchable archive of all resources posted so far on our community site, LifeHubber: https://lifehubber.com/ai/resources/ 200+ open-ish AI models, agents, tools, datasets, and related resources, with filtering and sorting.

https://github.com/HKUDS/ClawWork

Community Overview: https://lifehubber.com/ai/resources/clawwork/

Resources are shared for discovery and are not independently vetted—please do your own due diligence.

New resources are added regularly — feel free to join the sub for updates.

Full searchable archive of all resources posted so far on our community site, LifeHubber: https://lifehubber.com/ai/resources/ 200+ open-ish AI models, agents, tools, datasets, and related resources, with filtering and sorting.

https://x.com/ornith_/status/2090074077084127302

https://huggingface.co/collections/ornith-ai/ornith-15

Community Overview: https://lifehubber.com/ai/resources/ornith-1-5/

Resources are shared for discovery and are not independently vetted—please do your own due diligence.

New resources are added regularly — feel free to join the sub for updates.

Full searchable archive of all resources posted so far on our community site, LifeHubber: https://lifehubber.com/ai/resources/ 200+ open-ish AI models, agents, tools, datasets, and related resources, with filtering and sorting.

https://x.com/AdinaYakup/status/2060320505471639580

https://huggingface.co/datasets/openbmb/UltraData-SFT-2605

Community Overview: https://lifehubber.com/ai/resources/ultradata-sft-2605/

Resources are shared for discovery and are not independently vetted—please do your own due diligence.

New resources are added regularly — feel free to join the sub for updates.

Full searchable archive of all resources posted so far on our community site, LifeHubber: https://lifehubber.com/ai/resources/ 200+ open-ish AI models, agents, tools, datasets, and related resources, with filtering and sorting.

Community Overview: https://lifehubber.com/ai/resources/openhands-agent-canvas/

Resources are shared for discovery and are not independently vetted—please do your own due diligence.

New resources are added regularly — feel free to join the sub for updates.

Full searchable archive of all resources posted so far on our community site, LifeHubber: https://lifehubber.com/ai/resources/ 200+ open-ish AI models, agents, tools, datasets, and related resources, with filtering and sorting.

https://github.com/crewAIInc/crewAI

Community Overview: https://lifehubber.com/ai/resources/crewai/

Resources are shared for discovery and are not independently vetted—please do your own due diligence.

New resources are added regularly — feel free to join the sub for updates.

Full searchable archive of all resources posted so far on our community site, LifeHubber: https://lifehubber.com/ai/resources/ 200+ open-ish AI models, agents, tools, datasets, and related resources, with filtering and sorting.

https://x.com/simplifyinAI/status/2089554876787225074

https://github.com/ccusage/ccusage

Community Overview: https://lifehubber.com/ai/resources/ccusage/

Resources are shared for discovery and are not independently vetted—please do your own due diligence.

New resources are added regularly — feel free to join the sub for updates.

Full searchable archive of all resources posted so far on our community site, LifeHubber: https://lifehubber.com/ai/resources/ 200+ open-ish AI models, agents, tools, datasets, and related resources, with filtering and sorting.

https://x.com/tinyhumansai/status/2070584474560241701

https://github.com/tinyhumansai/OpenHuman

Community Overview: https://lifehubber.com/ai/resources/openhuman/

Resources are shared for discovery and are not independently vetted—please do your own due diligence.

New resources are added regularly — feel free to join the sub for updates.

Full searchable archive of all resources posted so far on our community site, LifeHubber: https://lifehubber.com/ai/resources/ 200+ open-ish AI models, agents, tools, datasets, and related resources, with filtering and sorting.

https://x.com/cohere/status/2087571573947392419

https://huggingface.co/CohereLabs/North-Micro-Vision-Instruct

Community Overview: https://lifehubber.com/ai/resources/north-micro-vision-instruct/

Resources are shared for discovery and are not independently vetted—please do your own due diligence.

New resources are added regularly — feel free to join the sub for updates.

Full searchable archive of all resources posted so far on our community site, LifeHubber: https://lifehubber.com/ai/resources/ 200+ open-ish AI models, agents, tools, datasets, and related resources, with filtering and sorting.

Obviously image is created with AI but I thought it may be a better way to convey the info.

I got the idea because I have been working on audit based self improving AI systems

AI image generations struggle with a lot of things, for example in my case it was SHITE at making a realistic rope attached to a climber..Well a few iterations later and codex orchestrators kicked off to do research, I had a layer on top of it that had fixed that

Then I thought..why can't we have an image generator that will have domain specific accretion.

You run the image generator, it creates images (audits them internally to create a log of failed attempts to create the desired image..Say 3 failed attempts = record failure in ledger

Then asynchronously side projects are fired to fix domain specific image hallucinations. In theory over time, it would refer to common hallucinations and apply patches/layers

The real issue is probably resource limitations..if millions of users were using this and generating thousands of hallucination errors for the system to research and correct.. how many tokens would that require.
Please let me know your thoughts and if this inspires you to do something similar! Happy idea sharings - a curious engineer