“Life is old there, older than the trees
Younger than the mountains”
- surprisingly scientifically accurate country road, 1971
Cohere connect week kicks off today in Toronto :)
genuinely so humbling to meet the whole team in person.
Some of the best people around.
Happy Canada Day everyone. This place rules. Let’s keep making it better :)
Guys I love football now 🇨🇦
Thanks for talking :)
He's raised over $1.5 billion in capital for his company, @cohere , most recently valued at over $6 billion — and he's one of the sharpest minds in AI today.
My conversation with @nickfrosst , co-founder of @cohere is live.
I got 60 minutes of his time for the new episode of Make It Click.
We got into:
– How Cohere is competing and winning in the enterprise
– How he spends his day at the office day
– The unit economics of B2B vs. B2C among LLM companies
And on top of it all, he's the lead singer of a band with 3.5M+ monthly listeners on Spotify.
Hope you enjoy!
Show more
now seems like a good day to remind people we have an apache 2.0 coding model you can run with 20 gigs of ram locally for free forever.
Its a 4 bit quant that gets >99% original performance.
Show more
This is bad for Canada. Broadening access to gambling isn't progress. Our tech industry should aspire to more.
Prediction markets are coming to Canada.
For the first time, millions of Canadians will be able to trade what's next through Wealthsimple.
new 4 bit quantization of north mini code. you can run this on anything with 20 gigs of memory, or for free on @opencode !
We just dropped three updates that make North Mini Code more accessible than ever ⬇️
Thanks for having me :)
On The BetaKit Podcast, @cohere's @nickfrosst talks sovereign AI, Star Trek, and why he doesn’t make a point of listening to Marc Andreessen.
We won’t
Canadian LLMs would never do this to you
Anthropic’s behaviour here makes it clear their motivation was their bottom line and not for “safety”
Open source and locally deployable models are the way.
NEW: Anthropic is walking back Claude Fable 5's policy to covertly degrade performance for competing AI researchers, after facing fierce backlash.
“We’re changing Fable 5’s safeguards for frontier LLM development to make them visible,” Anthropic tells WIRED. “We made the wrong tradeoff and we apologize for not getting the balance right.”
Show more
if you think data centers in space are a good idea, you are exit liquidity.
ill be joining the show in 10 minutes :)
COHERE CO-FOUNDER | FABLE REACTIONS | IS MYTHOS NERFED?
New leaderboard for audio transcription just launched and our apache 2.0 Cohere-Transcribe is at the top. This eval didn't exist when we trained the model, so its nice to see us do so well on it.
Show more
this model is the opposite of mythos.
Its small, cost effective, apache 2.0, and locally deployable. This is the way LLMs should go.
small, open source, transparent and sovereign
vs
large, expensive, proprietary and hegemonic
Show more
Introducing Cohere's first open-source coding model: North Mini Code
Small & efficient, designed for agentic performance and built for community input.
Cohere just released North Mini Code, a small 30B parameter (3B active) open weights coding model that scores 27.6 on the Artificial Analysis Intelligence Index
Less than a month since @cohere's last model release, Command A+, has launched another open weights model that is optimized for coding, and much smaller at 30B total parameters and 3B active parameters.
Key Takeaways:
➤ Achieves 27.6 on the Artificial Analysis Intelligence Index, above gpt-oss-20B (high) at 24.5 and just below Mistral Small 4 (119B parameters, 6.5B active) at 27.8
➤ Scores competitively on the Artificial Analysis Coding Index (weighted average of Terminal-Bench Hard and SciCode) against open weights models in its size class, scoring 33.4, significantly above GLM-4.7-Flash at 25.9, and below Qwen3.6 35B A3B at 35.2. However, it underperforms on non-coding agentic tasks, scoring 14% on GDPval-AA and 37% on 𝜏²-Bench Telecom
➤ On Cohere’s API, North Mini Code is faster than several comparable open weights models of its intelligence and size class (~199 output tokens per second)
➤ North Mini Code is a text-only 30B total parameter and 3B active parameter model, and is open-sourced under the Apache 2.0 license
Show more
We made a small coding model. Its open source apache 2.0.
Now more than ever i think this tech needs to be built in public so that those using it are in control. Try it out if you want a small and efficient coding model.
Show more
