Ministral 3
Paper
• 2601.08584 • Published
• 55
AI & ML interests
None defined yet.
Recent Activity
View all activity
1aurent
authored a paper 25 days ago
JingzeShi
authored 2 papers 27 days ago
JingzeShi
submitted a paper to Daily Papers 27 days ago
julien-c
submitted a paper to Daily Papers about 1 month ago
csabakecskemeti
posted an
update 2 months ago
Post
3280
Just sharing a result of a homelab infrastructure experiment:
I've managed to setup a distributed inference infra at home using a DGX Spark (128GB unified gddr6) and a linux workstation with an RTX 6000 Pro (96GB gddr7) connected via 100Gbps RoCEv2. The model I've used (https://lnkd.in/gx6J7YuB) is about 140GB so could not fit either of the GPU. Full setup and tutorial soon on devquasar.com
Screen recording:
https://lnkd.in/gKM9H5GJ
I've managed to setup a distributed inference infra at home using a DGX Spark (128GB unified gddr6) and a linux workstation with an RTX 6000 Pro (96GB gddr7) connected via 100Gbps RoCEv2. The model I've used (https://lnkd.in/gx6J7YuB) is about 140GB so could not fit either of the GPU. Full setup and tutorial soon on devquasar.com
Screen recording:
https://lnkd.in/gKM9H5GJ
-
- 3 replies
kenobi
authored 3 papers 3 months ago
On Invariance Penalties for Risk Minimization
Paper
• 2106.09777 • Published
Generating (Factual?) Narrative Summaries of RCTs: Experiments with Neural Multi-Document Summarization
Paper
• 2008.11293 • Published
Bayesian Deep Learning for Exoplanet Atmospheric Retrieval
Paper
• 1811.03390 • Published
woojun-jung
authored a paper 3 months ago
Aurelien-Morgan
posted an
update 3 months ago
Post
355
Hey, I went to Hangzhou to talk about
The recording just got released. Go check it out !
https://www.youtube.com/watch?v=nmrMachM5aM
Slides are there :
https://docs.google.com/presentation/d/1hnAzHJ0SbeAOtGJir-iH84RBtXT1OxVT/
retrain-pipelines at the GOSIM Foundation's conference last september.The recording just got released. Go check it out !
https://www.youtube.com/watch?v=nmrMachM5aM
Slides are there :
https://docs.google.com/presentation/d/1hnAzHJ0SbeAOtGJir-iH84RBtXT1OxVT/
-
- 2 replies
csabakecskemeti
posted an
update 3 months ago
Post
1350
FYI: Mistral.Ministral-3 dequantizer FP8->BF16
https://github.com/csabakecskemeti/ministral-3_dequantizer_fp8-bf16
(The instruct model weights are in FP8)
https://github.com/csabakecskemeti/ministral-3_dequantizer_fp8-bf16
(The instruct model weights are in FP8)
csabakecskemeti
posted an
update 3 months ago
Post
2094
Looking for some help to test an INT8 Deepseek 3.2:
SGLang supports Channel wise INT8 quants on CPUs with AMX instructions (Xeon 5 and above AFAIK)
https://lmsys.org/blog/2025-07-14-intel-xeon-optimization/
Currently uploading an INT8 version of Deepseek 3.2 Speciale:
DevQuasar/deepseek-ai.DeepSeek-V3.2-Speciale-Channel-INT8
I cannot test this I'm on AMD
"AssertionError: W8A8Int8LinearMethod on CPU requires that CPU has AMX support"
(I assumed it can fall back to some non optimized kernel but seems not)
If anyone with the required resources (Intel Xeon 5/6 + ~768-1TB ram) can help to test this that would be awesome.
If you have hints how to make this work on AMD Threadripper 7000 Pro series please guide me.
Thanks all!
SGLang supports Channel wise INT8 quants on CPUs with AMX instructions (Xeon 5 and above AFAIK)
https://lmsys.org/blog/2025-07-14-intel-xeon-optimization/
Currently uploading an INT8 version of Deepseek 3.2 Speciale:
DevQuasar/deepseek-ai.DeepSeek-V3.2-Speciale-Channel-INT8
I cannot test this I'm on AMD
"AssertionError: W8A8Int8LinearMethod on CPU requires that CPU has AMX support"
(I assumed it can fall back to some non optimized kernel but seems not)
If anyone with the required resources (Intel Xeon 5/6 + ~768-1TB ram) can help to test this that would be awesome.
If you have hints how to make this work on AMD Threadripper 7000 Pro series please guide me.
Thanks all!
-
- 8 replies
csabakecskemeti
posted an
update 4 months ago
Post
313
Recently there are so much activity on token efficient formats, I've also build a package (inspired by toon).
Deep-TOON
My goal was to token efficiently handle json structures with complex embeddings.
So this is what I've built on the weekend. Feel free try:
https://pypi.org/project/deep-toon/0.1.0/
Deep-TOON
My goal was to token efficiently handle json structures with complex embeddings.
So this is what I've built on the weekend. Feel free try:
https://pypi.org/project/deep-toon/0.1.0/
csabakecskemeti
posted an
update 5 months ago
Post
2654
Christmas came early this year
- 3 replies
mbrack
authored a paper 5 months ago
Post
8110
We're kick-starting the process of Transformers v5, with @ArthurZ and @cyrilvallez !
v5 should be significant: we're using it as a milestone for performance optimizations, saner defaults, and a much cleaner code base worthy of 2025.
Fun fact: v4.0.0-rc-1 came out on Nov 19, 2020, nearly five years ago!
v5 should be significant: we're using it as a milestone for performance optimizations, saner defaults, and a much cleaner code base worthy of 2025.
Fun fact: v4.0.0-rc-1 came out on Nov 19, 2020, nearly five years ago!
- 6 replies
teknium
authored 2 papers 6 months ago
jeffboudier
posted an
update 6 months ago
Post
3206
Quick 30s demo of the new Hub > Azure AI integration to deploy HF models in your own Azure account. Now with Py and CLI!
GG @alvarobartt @kramp @pagezyhf
GG @alvarobartt @kramp @pagezyhf