Hugging Face
Models
Datasets
Spaces
Buckets
new
Docs
Enterprise
Pricing
Website
Tasks
HuggingChat
Collections
Languages
Organizations
Community
Blog
Posts
Daily Papers
Hardware
Learn
Discord
Forum
GitHub
Solutions
Team & Enterprise
Hugging Face PRO
Enterprise Support
Inference Providers
Inference Endpoints
Storage Buckets
Log In
Sign Up
thinletter
community
https://thinletter.io
Activity Feed
Follow
1
AI & ML interests
None defined yet.
Recent Activity
honza-rosecky
updated
a Space
about 9 hours ago
thinletter/README
honza-rosecky
updated
a model
about 9 hours ago
thinletter/ThinCtx-Qwen3.5-2B
honza-rosecky
published
a model
about 9 hours ago
thinletter/ThinCtx-Qwen3.5-2B
View all activity
Team members
1
thinletter
's activity
All
Models
Datasets
Spaces
Buckets
Papers
Collections
Community
Posts
Articles
honza-rosecky
updated
a Space
about 9 hours ago
Running
thinletter
🪶
honza-rosecky
updated
a model
about 9 hours ago
thinletter/ThinCtx-Qwen3.5-2B
Text Generation
•
2B
•
Updated
about 7 hours ago
•
1
honza-rosecky
published
a model
about 9 hours ago
thinletter/ThinCtx-Qwen3.5-2B
Text Generation
•
2B
•
Updated
about 7 hours ago
•
1
honza-rosecky
updated
a model
about 9 hours ago
thinletter/ThinCtx-Qwen3.5-0.8B
Text Generation
•
0.8B
•
Updated
about 7 hours ago
•
1
honza-rosecky
published
a model
about 9 hours ago
thinletter/ThinCtx-Qwen3.5-0.8B
Text Generation
•
0.8B
•
Updated
about 7 hours ago
•
1
honza-rosecky
updated
2 models
22 days ago
thinletter/qwen3-embedding-0.6b-vq-clients
Updated
22 days ago
•
8
thinletter/harrier-0.6b-vq-clients
Feature Extraction
•
Updated
22 days ago
•
107
honza-rosecky
updated
a Space
24 days ago
Running
thinletter demo
🔎
Search scientific papers locally with a browser‑based query encoder
honza-rosecky
in
thinletter/harrier-0.6b-vq-clients
27 days ago
Prompt-K/V variants added: four files calibrated with the query prompt's keys and values in full precision
#5 opened 27 days ago by
honza-rosecky
honza-rosecky
in
thinletter/qwen3-embedding-0.6b-vq-clients
27 days ago
First Czech vector-quantised client: 237 MiB, 96.8 % of fp32 nDCG@10 on WebFAQ-cs, runs in the browser
#1 opened 27 days ago by
honza-rosecky
honza-rosecky
published
a model
27 days ago
thinletter/qwen3-embedding-0.6b-vq-clients
Updated
22 days ago
•
8
honza-rosecky
published
a Space
27 days ago
Running
thinletter demo
🔎
Search scientific papers locally with a browser‑based query encoder
honza-rosecky
updated
3 models
27 days ago
thinletter/bge-m3-query-clients
Feature Extraction
•
0.6B
•
Updated
27 days ago
•
111
thinletter/qwen3-embedding-0.6b-query-clients
Feature Extraction
•
0.6B
•
Updated
27 days ago
•
200
thinletter/harrier-0.6b-query-clients
Feature Extraction
•
0.6B
•
Updated
27 days ago
•
110
honza-rosecky
in
thinletter/harrier-0.6b-vq-clients
27 days ago
The runtime is now open source: quantiser, .vqw compiler and WebGPU runtime (Apache-2.0)
#4 opened 27 days ago by
honza-rosecky
honza-rosecky
in
thinletter/harrier-0.6b-vq-clients
about 1 month ago
The query prompt's K/V in full precision: +0.01–0.04 nDCG@10 at 1.6–1.8 bits per weight for 2 MB of data (measured, not yet in the files)
2
#3 opened about 1 month ago by
honza-rosecky
honza-rosecky
published
a Space
about 1 month ago
Running
thinletter
🪶
honza-rosecky
in
thinletter/harrier-0.6b-vq-clients
about 1 month ago
Vector-quantised harrier-0.6b query clients at 1.8–2.1 bits per weight (105–127 MiB) — summary and links
#2 opened about 1 month ago by
honza-rosecky
honza-rosecky
in
thinletter/harrier-0.6b-query-clients
about 1 month ago
harrier-0.6b query clients: 192–235 MiB, 94–99 % of fp32 on the unchanged index — summary and links
#1 opened about 1 month ago by
honza-rosecky
Load more