paulplee commited on
Commit
67d3729
·
verified ·
1 Parent(s): 378d39a

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +17 -0
README.md CHANGED
@@ -11,7 +11,14 @@ tags:
11
  - homelab
12
  - open-data
13
  - consumer-hardware
 
 
 
 
 
14
  pretty_name: Poor Paul's Benchmark Results
 
 
15
  size_categories:
16
  - 1K<n<10K
17
  ---
@@ -43,6 +50,16 @@ Current benchmark runners represented in the dataset include:
43
  - `llama-server` — TTFT and ITL latency with real conversational prompts
44
  - `llama-server-loadtest` — max concurrency discovery and concurrency curves
45
 
 
 
 
 
 
 
 
 
 
 
46
  ## Dataset Structure
47
 
48
  Each row corresponds to one normalized benchmark result.
 
11
  - homelab
12
  - open-data
13
  - consumer-hardware
14
+ - poor-pauls-benchmark
15
+ - gpu-benchmark
16
+ - quantization
17
+ - vram
18
+ - mcp
19
  pretty_name: Poor Paul's Benchmark Results
20
+ language:
21
+ - en
22
  size_categories:
23
  - 1K<n<10K
24
  ---
 
50
  - `llama-server` — TTFT and ITL latency with real conversational prompts
51
  - `llama-server-loadtest` — max concurrency discovery and concurrency curves
52
 
53
+ ## Ecosystem
54
+
55
+ | Component | Link |
56
+ |---|---|
57
+ | Benchmark tool | [poor-pauls-benchmark](https://github.com/paulplee/poor-pauls-benchmark) |
58
+ | MCP server | [ppb-mcp](https://github.com/paulplee/ppb-mcp) — lets LLMs query this dataset directly |
59
+ | Analytics dashboard | [poorpaul.dev/insights](https://poorpaul.dev/insights) |
60
+
61
+ Ask an LLM to query this dataset: connect any MCP client to `mcp.poorpaul.dev`.
62
+
63
  ## Dataset Structure
64
 
65
  Each row corresponds to one normalized benchmark result.