www.minds.com/jelyazko/
21world
AI & ML interests
(new) sentence at once = token. PRECOMPUTED AI WEIGHTS
! INNOVATIONS ARE USED INSTEAD REFUSALS !
Known Training Algos:
- Backpropagation
- Forward Forward
- Predictive Coding
- Q-Learning ?
-....
|for correct ds1307 pinout 8-stars(0-14B models Q4)| for Diffusion LLM 9-stars | Teach llms on ASCII/Text Based STL Modeling , ic pinouts .optionaly:patents. |context editing| .exact size of all obj dimentions. Virtualization of ai/ml in mcu .Funding require.///until llms are weak///
https://github.com/sponsors/jelspace
https://ko-fi.com/21world
https://rumble.com/c/c-5865836
https://www.buymeacoffee.com/21world
Recent Activity
updated
a collection
about 12 hours ago
18\ other models
upvoted
a
paper
about 14 hours ago
LangSplatV2: High-dimensional 3D Language Gaussian Splatting with 450+
FPS
updated
a collection
about 19 hours ago
18\ other models
Organizations
62\ Audio to Synchronous Video creation
49\ Video to synchronous Audio creation
60\ Video with Audio Creation
58\ VIDEO EDITOR
56\ -MOE- 256_experts - Text to Text
54\ Video Tools
52\ 28 layers X 1536 neurons [****______]
50\ Colorization
48\ TEXT -> 3D
46\ voice llm
44\ 24 layers(pcb) X 2048(?h_dim?) neurons [*_________]
Model quality test - 1/10 stars
42\ in static.video -> out text
40\ IN TEXT -> OUT TEXT , language translation.
38\ in text,static video -> out static video |for cpu usage
36\ in text -> out speech
34\ speech <-> text translate universal
32\ video creation
30\ Interesting.what is this ? how it works?
28\ 48 layers X 4096 (?dim?)neurons [*****_____]
Model quality test - 5/10 stars
26\ 32 layers X 2560 (?dim?)neurons [__________]
Model quality test - 0/10 stars
24\ 32 layers X 4096 (?dim?)neurons [*****_____]
Model quality test 5/10 stars
22\ who is who
20\ 80 layers X 8192 (?dim?)neurons [*******___]
Model quality test - 7/10 stars
18\ other models
shift counting to know who disappear from collection 1. alchemonaut/QuartetAnemoi-70B-t0.0001
16\ Strange thinking
Caution / avoid possible bad/unwanted media content , may useful for merging
14\ 40 layers X 6144 neurons [*_________]
Model quality chat test - 1/10 stars
12\ 48 layers X 4096 neurons [**________]
Model quality chat test - 2/10 stars
10\ 32 layers X 3072 neurons [***_______]
Model quality chat test - 3/10 stars
8\ 27 layers X 2048 neurons [*******___]
Model quality chat test (coding) - 7/10 stars
6\ 32 layers X 4096 neurons [*****_____]
Model quality chat test (coding) - 5/10 stars
4\ 2d to 3d - Video
2\ ?2 layers X ?2048 neurons X?? boards 8/64ex.[***'______]
Model quality chat test - 3.4/10 stars
63\ Document tools
61\ video iq /real skynet/
59\ -MOE- diffusion , pictures generation
57\ Picture Editors
55\ -MOE- Vision to Text
53\ Claimed usage of (?layer?precomputing) lookup table
not tested / code not checked
51\ Mask type TEXT | IMAGE LLM
Concept
49\ Video to synchronous Audio creation
62\ Audio to Synchronous Video creation
47\ Superb to amazing
45\ AI SIMULATOR
43\ in text/static.video -> out static.video - LCM - MOD
Additional Conceptual MOD
41\ IN TEXT -> OUT AUDIO
https://app.suno.ai/song/090c77cb-8fb1-4669-aa51-7e414eaed612
39\ UPSCALE RESOLUTION
37\ Sd model weights (ai data logic for static.video gen..)
35\ in speech -> out text
-
distil-whisper/distil-large-v2
Automatic Speech Recognition • 0.8B • Updated • 10.8k • 510 -
openai/whisper-large-v3
Automatic Speech Recognition • 2B • Updated • 2.96M • • 4.63k -
ggerganov/whisper.cpp
Automatic Speech Recognition • Updated • 1.05k -
lmms-lab/Aero-1-Audio
Text Generation • 2B • Updated • 118 • 82
33\ |.c ||.cpp||
https://github.com/foldl/chatllm.cpp/blob/master/docs/fun.md
31\ LLM Best Models
-
Running on CPU Upgrade13.3k13.3k
Open LLM Leaderboard
🏆Track, rank and evaluate open LLMs and chatbots
-
Running1.37k1.37k
Big Code Models Leaderboard
📈Search and submit code models for evaluation
-
Running4.52k4.52k
Chatbot Arena Leaderboard
🏆View chatbot performance leaderboard
-
Running on CPU Upgrade820820
Open VLM Leaderboard
🌎VLMEvalKit Evaluation Results Collection
29\ OK
27\ 96 layers X 8192 (?dim?)neurons [*******___]
Model quality test - 7/10 stars
25\ 48 layers X 8192 (?dim?)neurons [******____]
Model quality test - 6
/10 stars
23\ Info - pages
1. TinyGSM: achieving >80% on GSM8k with small language models
-
TinyGSM: achieving >80% on GSM8k with small language models
Paper • 2312.09241 • Published • 39 -
ShortGPT: Layers in Large Language Models are More Redundant Than You Expect
Paper • 2403.03853 • Published • 66 -
Gamba: Marry Gaussian Splatting with Mamba for single view 3D reconstruction
Paper • 2403.18795 • Published • 21 -
Diffusion-RWKV: Scaling RWKV-Like Architectures for Diffusion Models
Paper • 2404.04478 • Published • 13
21\ 60 layers X 7168 (?dim?)neurons [******____]
Model quality test - 6 / 10 stars
19\ 22 layers X 2048(?h_dim?) neurons [*_________]
Model quality test - 1/10 stars
17\ ABSOLUT Perfect Bulgarian !
15\ 48 layers X 4096 neurons [*_________]
Model quality chat test - 1/10 stars
13\ 40 layers X 6144 neurons [***_______]
Model quality chat test 3/10 stars
11\ 32 layers X 2560 neurons [***_______]
Model quality chat test - 3/10 stars
9\ 32 layers X 4096 neurons [******____]
https://github.com/jameswdelancey/llama3.c Model quality test 6/10 stars
7\ Video observers
5\ 2D to 3D
3\ 2D to 3D CAD
/1\ Dataset
64\ [*** 6 Stars ***____]
63\ Document tools
62\ Audio to Synchronous Video creation
49\ Video to synchronous Audio creation
61\ video iq /real skynet/
60\ Video with Audio Creation
59\ -MOE- diffusion , pictures generation
58\ VIDEO EDITOR
57\ Picture Editors
56\ -MOE- 256_experts - Text to Text
55\ -MOE- Vision to Text
54\ Video Tools
53\ Claimed usage of (?layer?precomputing) lookup table
not tested / code not checked
52\ 28 layers X 1536 neurons [****______]
51\ Mask type TEXT | IMAGE LLM
Concept
50\ Colorization
49\ Video to synchronous Audio creation
62\ Audio to Synchronous Video creation
48\ TEXT -> 3D
47\ Superb to amazing
46\ voice llm
45\ AI SIMULATOR
44\ 24 layers(pcb) X 2048(?h_dim?) neurons [*_________]
Model quality test - 1/10 stars
43\ in text/static.video -> out static.video - LCM - MOD
Additional Conceptual MOD
42\ in static.video -> out text
41\ IN TEXT -> OUT AUDIO
https://app.suno.ai/song/090c77cb-8fb1-4669-aa51-7e414eaed612
40\ IN TEXT -> OUT TEXT , language translation.
39\ UPSCALE RESOLUTION
38\ in text,static video -> out static video |for cpu usage
37\ Sd model weights (ai data logic for static.video gen..)
36\ in text -> out speech
35\ in speech -> out text
-
distil-whisper/distil-large-v2
Automatic Speech Recognition • 0.8B • Updated • 10.8k • 510 -
openai/whisper-large-v3
Automatic Speech Recognition • 2B • Updated • 2.96M • • 4.63k -
ggerganov/whisper.cpp
Automatic Speech Recognition • Updated • 1.05k -
lmms-lab/Aero-1-Audio
Text Generation • 2B • Updated • 118 • 82
34\ speech <-> text translate universal
33\ |.c ||.cpp||
https://github.com/foldl/chatllm.cpp/blob/master/docs/fun.md
32\ video creation
31\ LLM Best Models
-
Running on CPU Upgrade13.3k13.3k
Open LLM Leaderboard
🏆Track, rank and evaluate open LLMs and chatbots
-
Running1.37k1.37k
Big Code Models Leaderboard
📈Search and submit code models for evaluation
-
Running4.52k4.52k
Chatbot Arena Leaderboard
🏆View chatbot performance leaderboard
-
Running on CPU Upgrade820820
Open VLM Leaderboard
🌎VLMEvalKit Evaluation Results Collection
30\ Interesting.what is this ? how it works?
29\ OK
28\ 48 layers X 4096 (?dim?)neurons [*****_____]
Model quality test - 5/10 stars
27\ 96 layers X 8192 (?dim?)neurons [*******___]
Model quality test - 7/10 stars
26\ 32 layers X 2560 (?dim?)neurons [__________]
Model quality test - 0/10 stars
25\ 48 layers X 8192 (?dim?)neurons [******____]
Model quality test - 6
/10 stars
24\ 32 layers X 4096 (?dim?)neurons [*****_____]
Model quality test 5/10 stars
23\ Info - pages
1. TinyGSM: achieving >80% on GSM8k with small language models
-
TinyGSM: achieving >80% on GSM8k with small language models
Paper • 2312.09241 • Published • 39 -
ShortGPT: Layers in Large Language Models are More Redundant Than You Expect
Paper • 2403.03853 • Published • 66 -
Gamba: Marry Gaussian Splatting with Mamba for single view 3D reconstruction
Paper • 2403.18795 • Published • 21 -
Diffusion-RWKV: Scaling RWKV-Like Architectures for Diffusion Models
Paper • 2404.04478 • Published • 13
22\ who is who
21\ 60 layers X 7168 (?dim?)neurons [******____]
Model quality test - 6 / 10 stars
20\ 80 layers X 8192 (?dim?)neurons [*******___]
Model quality test - 7/10 stars
19\ 22 layers X 2048(?h_dim?) neurons [*_________]
Model quality test - 1/10 stars
18\ other models
shift counting to know who disappear from collection 1. alchemonaut/QuartetAnemoi-70B-t0.0001
17\ ABSOLUT Perfect Bulgarian !
16\ Strange thinking
Caution / avoid possible bad/unwanted media content , may useful for merging
15\ 48 layers X 4096 neurons [*_________]
Model quality chat test - 1/10 stars
14\ 40 layers X 6144 neurons [*_________]
Model quality chat test - 1/10 stars
13\ 40 layers X 6144 neurons [***_______]
Model quality chat test 3/10 stars
12\ 48 layers X 4096 neurons [**________]
Model quality chat test - 2/10 stars
11\ 32 layers X 2560 neurons [***_______]
Model quality chat test - 3/10 stars
10\ 32 layers X 3072 neurons [***_______]
Model quality chat test - 3/10 stars
9\ 32 layers X 4096 neurons [******____]
https://github.com/jameswdelancey/llama3.c Model quality test 6/10 stars
8\ 27 layers X 2048 neurons [*******___]
Model quality chat test (coding) - 7/10 stars
7\ Video observers
6\ 32 layers X 4096 neurons [*****_____]
Model quality chat test (coding) - 5/10 stars
5\ 2D to 3D
4\ 2d to 3d - Video
3\ 2D to 3D CAD
2\ ?2 layers X ?2048 neurons X?? boards 8/64ex.[***'______]
Model quality chat test - 3.4/10 stars
/1\ Dataset