mistralai/Magistral-Small-2506
Text Generation
•
Updated
•
11.6k
•
•
402
Mindkrypted
mindkrypted
AI & ML interests
*.*
Recent Activity
liked
a model
4 days ago
mistralai/Magistral-Small-2506
liked
a model
16 days ago
turboderp/Llama-3.1-Nemotron-Ultra-253B-v1-exl3
liked
a model
about 1 month ago
calcuis/wan-gguf
Organizations
mindkrypted's activity
liked
a
model
4 days ago
liked
a
model
16 days ago
liked
a
model
about 1 month ago
liked
3
models
about 2 months ago
liked
a
model
2 months ago
upvoted
an
article
2 months ago
Article
Comparing sub 50GB Llama 4 Scout quants (KLD/Top P)
By
•
•
40
liked
a
model
2 months ago
reacted to
sr-rai's
post with 🤗
2 months ago
Post
2760
ExLlamaV3 is out. And it introduces EXL3 - a new SOTA quantization format!
"The conversion process is designed to be simple and efficient and requires only an input model (in HF format) and a target bitrate. By computing Hessians on the fly and thanks to a fused Viterbi kernel, the quantizer can convert a model in a single step, taking a couple of minutes for smaller models, up to a few hours for larger ones (70B+) (on a single RTX 4090 or equivalent GPU.)"
Repo: https://github.com/turboderp-org/exllamav3
"The conversion process is designed to be simple and efficient and requires only an input model (in HF format) and a target bitrate. By computing Hessians on the fly and thanks to a fused Viterbi kernel, the quantizer can convert a model in a single step, taking a couple of minutes for smaller models, up to a few hours for larger ones (70B+) (on a single RTX 4090 or equivalent GPU.)"
Repo: https://github.com/turboderp-org/exllamav3
- 1 reply
upvoted
a
collection
2 months ago
liked
a
model
3 months ago
Fix required for config.json >> BOS Token ID - Incoherent outputs
1
#8 opened 5 months ago
by
mindkrypted
liked
a
model
4 months ago
liked
a
Space
4 months ago
liked
a
model
4 months ago
liked
a
model
5 months ago
