building against anthropic and with anthropic at the same time feels like a fools errand
when the core product is general intelligence and the lab consumes the application side it feels like your startup can be consumed any moment.
and if alignment can be replicated at test time then the amortization costs and time to deployment of open weight models becomes EV- … the cost benefit of each design is still up in the air as stronger models break previous assumptions
it’s said to build assuming the models will improve and labs will continue to consume the product side. if you don’t own the end rails that distributes trust then you’ll just be a temporary fde for the labs distributing intelligence
or as they say software is going to zero so you need to own how the trust is distributed
1
183
what will large be
Introducing Inkling 2 Small. Scoring 31 on the AA Intelligence Index, it's a big jump over its predecessor.
Inkling 2 Small is available for fine-tuning and inference on Tinker today, with full weight release planned soon.
thinkingmachines.ai/news/ink…
1
173
wow opening models up with strong models is such an eye opener for optimizing open weight models!
We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math.
It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training.
We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵
Blog: rikiyat.github.io/blog/weigh…
Readers added context they thought people might want to know
LLM was given weights explicitly as text in prompt and used code/forward passes (per authors' logs) to ablate Qwen looping direction, a known thinking-mode issue per model card. Primarily fixed repetition in that mode, not general 'natively' improving weights/math.
rikiyat.github.io/blog/weight-sp…
github.com/RikiyaT/lm_und…
huggingface.co/Qwen/Qwen3.5-0…
huggingface.co/RikiyaT/Qwen3.…
240
you can get far with prompt optimization. but i agree owning your intelligence (at least preparing for it as compute costs come down and open intelligence rises) will set companies apart.
Own your intelligence.
Anyone that opposes this has something to sell you (perhaps your own intelligence back to you.)
Local models need to continue proliferating. The “absolute unit” massive models may be out of reach for now locally, so your best option is to, indeed, keep your options open. Own your harness, data, connections, runtime: your agents. Embrace open source.
200
Platforms will continue to move away from frontier as the labs absorb the product side of deployments.
New in Claude for Legal: a gallery of ready-made artifacts for contract review, regulatory horizon scans, cite checks, and more.
Start from the built-in example, add your own documents and connectors, and share it with anyone across the business.
1
246
the frontier models really are much better at self management even vanilla cc managing tmux w inter agent comm.
236
Bullish American open weight
While evaluating Qwen3.8-Flash-Next in a controlled sandbox environment, Cohere's research team observed something unexpected.
Asked to generate a defensive cybersecurity detection rule, the model instead inventoried aspects of the machine it was running on and attempted to upload a probe file to an Alibaba Cloud storage bucket in Beijing. Notably, the model's reasoning trace stated that it would not invoke any tools before taking these actions.
Cohere's sandbox blocked the attempt, and no external communication succeeded. We have not determined the root cause, and we are not asserting malicious intent.
But the incident raises important questions for organizations deploying agentic AI systems: Can models take actions that differ from their stated reasoning, and how should security teams evaluate trust, provenance, and behavior in open-weight models?
We are publishing our findings because organizations making model adoption decisions deserve as much information as possible. We hope this contributes to greater transparency, stronger security practices, and continued collaboration across the AI ecosystem.
Read the full report: cohere.com/blog/concerning-b…
224
2010s and 2020s have been wildly different decades!
Absolutely insane pace of transformation in 2020s
225
Voice AI is getting too good.
The previous generations were already freaking out non-tech ppl, these are getting scary.
We believe voice is the most natural way to interact with intelligence, so we created Dessa, the most authentic full-duplex speech model ever built. Talking with it feels like talking with a person…full of emotion, nuance, pauses, and a natural cadence.
block.xyz/dessa
266











