Lample: Mistral is far from reaching pre-training saturation
Guillaume Lample · Mistral: Voxtral TTS, Forge, Leanstral, & Mistral 4 — w/ Pavan Kumar Reddy & Guillaume Lample · Mar 30, 2026 · at 45:23
Mistral AI Chief Scientist Guillaume Lample responds to industry debates regarding whether frontier labs should shift the majority of training compute away from pre-training toward reinforcement learning.
0:00 / 0:04exact quote · 4.6s
▶ Watch the full episode on YouTube → 720p mp4 · rendered on demand · StarZero watermark
“We are still working a lot on the pre-training side. We are very, very far from any sort of situation on the pre-training.”
quote is from the automated transcript, cleaned for reading: filler sounds and stutters are removed, nothing is rephrased. names can be misheard (the analysis reads context, assessments check outside sources). how →
More from Guillaume Lample
Lample: Mistral plans a phased approach to full-duplex audio models
“Ultimately what we want to do is to be this, Full duplex model, but we are not going to start this, start there directly. I think it's some approach that people are doing, but. Just to confirm, full duplex means it can speak while I'm speaking, or? Okay. Yeah,…”
Lample: Specialized AI models are more cost-effective than monolithic models
“That's why we can actually use models audio, but also like OCRs that are like really, really good at that, and that will be much more cost effective than a general model. That will contain a lot of capabilities you don't really need.”
Lample: 1B to 3B parameter models are optimal for speech transcription
“For instance, for audio here, if you want to do transcription, I think it makes no sense to use a model as this large. If you just want to transcribe tech, it would be very inefficient. Like if you want to do audio, you probably just want to do the one B or a …”
Lample: Long-horizon RL trajectories require new algorithms beyond GRPO
“GRPO, for instance, it doesn't really work with any bit of policy, which was okay initially, because you are solving math problems that can be solved in like a few thousand tokens, so the model can actually generate them pretty quickly, so when you do your upd…”
Lample: Voxtral TTS matches leading models at a fraction of cost
“So we support nine languages and this is a pretty small model a three-dimensional model, so very fast, and also state-ups, yeah, very equal. Performed at the same level of the best model, but it's Much more efficient in terms of cost, and also much, in terms o…”
Lample: Post-training breakthroughs like DPO were impossible without open LLaMA
“And if you look at many of the techniques that were developed after, for instance, Temma was open source, like all these post-training approaches like even DPOD, like performance optimization, all of this were done by people that had access to this model, and …”