Pablo Suarez commited on
Commit
fc065f9
·
unverified ·
1 Parent(s): a619d03

Revise README formatting and content

Browse files

Updated the README to include new formatting and additional context for the KV Cache Coordination Layer.

Files changed (1) hide show
  1. README.md +38 -21
README.md CHANGED
@@ -1,26 +1,43 @@
1
  # APOHARA V1.0 — ContextForge
2
 
3
  ```
4
- ╔══════════════════════════════════════════════════════════════════════════════╗
5
- ║ ║
6
- ║ ██████╗ ██╗ ██████╗ ██████╗ ██████╗ ██╗ ██╗███████╗███████╗███████╗ ║
7
- ║ ██╔════╝ ██║ ██╔═══██╗██╔══██╗██╔══██╗ ██║ ██║██╔════╝██╔════╝██╔════╝ ║
8
- ║ ██║ ███╗██║ ██║ ██║██████╔╝██████╔╝ ███████║█████╗ █████╗ ███████╗ ║
9
- ║ ██║ ██║██║ ██║ ██║██╔══██╗██╔══██╗ ██╔══██║██╔══╝ ██╔══╝ ╚════██║ ║
10
- ║ ╚██████╔╝█████████████╔╝██████╔╝██████╔╝ ██║ █████████╗██████╗███████║ ║
11
- ║ ╚═════╝ ══════╝ ╚════╝ ╚═════╝ ╚═════╝ ╚═╝ ╝╚══════╝╚══════╝╚══════╝ ║
12
-
13
- ║ ████████╗██████╗ █████████████████████████╗███████╗ █████╗ ██████╗
14
- ║ ╚══██╔══╝██╔══████╔══████╔════╝██╔════╝ ██╔════╝██╔════╝██╔══████╔══██
15
- ║ ██║ ██████╔███████║██║ █████╗ █████╗ ███████╗███████║██████╔
16
- ║ ██║ ██╔══██╗██╔══██║██║ ██╔══╝ ██╔══╝ ╚════██║██╔══██║██╔══██╗║
17
- ║ ██║ ██████╔╝████║╚██████╗███████╗ ███████╗████████████║ ██║║
18
- ║ ╚═╝ ╚═════╝ ╚═╝ ╝ ╚═══════════╝ ╚════════════╝╚ ╚═╝╚╝ ╚═╝
19
-
20
- ║ KV Cache Coordination Layer for Multi-Agent LLM Pipelines
21
- AMD Instinct MI300X · ROCm 7.x · HBM3 192 GB
22
- ║ ║
23
- ╚══════════════════════════════════════════════════════════════════════════════╝
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
24
  ```
25
 
26
  **Silicon-native KV cache coordination for multi-agent LLM pipelines on AMD Instinct MI300X**
@@ -527,4 +544,4 @@ Apache 2.0 — chosen for its patent protection and corporate adoption. GPL woul
527
  - RotateKV authors — *Pre-RoPE KV Quantization* (IJCAI 2025, arXiv:2501.16383)
528
  - vLLM-Omni authors — *Disaggregated Multimodal Serving* (Feb 2026, arXiv:2602.02204)
529
  - **Qwen team** — Qwen3-Embedding-0.6B and Qwen3.6-35B-A22B model availability on AMD ROCm
530
- - **LabLab.ai** — Hackathon platform and community
 
1
  # APOHARA V1.0 — ContextForge
2
 
3
  ```
4
+ # ▐▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▌
5
+ # ▐ ▌
6
+ # ▐ ▌
7
+ # ▐ ▌
8
+ # ▐ ▌
9
+ # ▐ ▌
10
+ # ▐ ███████████╗ ████████ ███████╗ ██████╗ █████╗ ▌
11
+ # ▐ ██╔══██╗██╔══██╗██╔═══██╗██║ ██║██╔══██╗██╔══██╗██╔══██╗ ▌
12
+ # ▐ █████████████╔╝██ ██║███████║███████║██████╔╝███████║ ▌
13
+ # ▐ ██╔══████╔═══╝ ██████╔══████╔══████╔══██╗██╔══██║
14
+ # ▐ ██████║ ╚██████╔╝██████████████║ ██║ ▌
15
+ # ▐ ╚═ ��═╝╚═╝ ╚═════╝ ╚═╝ ╚═╚═╝ ╚═╝╚═╝ ╚═╝╚═╝ ╚═╝ ▌
16
+ # ▐ ▌
17
+ # ▐ ███████████████╗███████╗███████╗████████████
18
+ # ██╔════╝██╔═══██╗████╗ ██║╚══██╔══╝██╔════╝╚██╗██╔╝╚══██╔
19
+ # ▐ ██ ██ ██║██╔██╗ ██║ ██║ █████╗ ╚███╔╝ ██║ ▌
20
+ # ▐ ██ ██║ ██║██║╚██╗██║ ██║ ██╔══╝ ██╔██╗ ██║ ▌
21
+ # ▐ ╚██████╗╚██████╔╝██╚████║ ██║ ███████╗██╔╝ ██╗ ██
22
+ # ▐ ╚═════╝ ╚═════╝ ╚═╝ ╚═══╝ ╚═╝ ╚══════╝╚═╝ ╚═╝ ╚═╝ ▌
23
+ # ▐ ▌
24
+ # ▐ ███████╗ ██████╗ ██████╗ ██████╗ ███████╗ ▌
25
+ # ▐ ██╔════╝██╔═══██╗██╔══██╗██╔════╝ ██╔════╝ ▌
26
+ # ▐ █████╗ ██║ ██║██████╔╝██║ ███╗█████╗ ▌
27
+ # ▐ ██╔══╝ ██║ ██║██╔══██╗██║ ██║██╔══╝ ▌
28
+ # ▐ ██║ ╚██████╔╝██║ ██║╚██████╔╝███████╗ ▌
29
+ # ▐ ╚═╝ ╚═════╝ ╚═╝ ╚═╝ ╚═════╝ ╚══════╝ ▌
30
+ # ▐ ▌
31
+ # ▐ ▌
32
+ # ▐ ▌
33
+ # ▐ KV Cache Coordination Layer for Multi-Agent LLM Pipelines ▌
34
+ # ▐ AMD Instinct MI300X · ROCm 7.x · HBM3 192 GB ▌
35
+ # ▐ ▌
36
+ # ▐ ▌
37
+ # ▐ ▌
38
+ # ▐▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▌
39
+
40
+
41
  ```
42
 
43
  **Silicon-native KV cache coordination for multi-agent LLM pipelines on AMD Instinct MI300X**
 
544
  - RotateKV authors — *Pre-RoPE KV Quantization* (IJCAI 2025, arXiv:2501.16383)
545
  - vLLM-Omni authors — *Disaggregated Multimodal Serving* (Feb 2026, arXiv:2602.02204)
546
  - **Qwen team** — Qwen3-Embedding-0.6B and Qwen3.6-35B-A22B model availability on AMD ROCm
547
+ - **LabLab.ai** — Hackathon platform and community