ai-speedometer 1.2.5 → 1.3.2

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/README.md CHANGED
@@ -2,6 +2,8 @@
2
2
 
3
3
  A CLI tool for benchmarking AI models across multiple providers with parallel execution and performance metrics.
4
4
 
5
+ ![Ai-speedometer showcase](./pics/image.png)
6
+
5
7
  ## Install
6
8
 
7
9
  ```bash
@@ -15,6 +17,12 @@ npm install -g ai-speedometer
15
17
  - **Tokens/Second** - Real-time throughput
16
18
  - **Token Counts** - Input, output, and total tokens used
17
19
 
20
+ ## New Features
21
+
22
+ - **REST API Default** - REST API benchmarking is now the default method for better compatibility
23
+ - **Headless Mode** - Run benchmarks without interactive CLI using command-line arguments
24
+ - **Streaming Support** - Full streaming support now available in REST API benchmarks
25
+
18
26
  ## Quick Setup
19
27
 
20
28
  1. **Set Model**
@@ -39,7 +47,8 @@ npm install -g ai-speedometer
39
47
  4. **Run Benchmark**
40
48
  ```bash
41
49
  ai-speedometer
42
- # Select "Run Benchmark (AI SDK)" → Choose models → Press ENTER
50
+ # Select "Run Benchmark (REST API)" → Choose models → Press ENTER
51
+ # Note: REST API is now the default benchmark method
43
52
  ```
44
53
 
45
54
  ## Usage
@@ -53,6 +62,13 @@ aispeed
53
62
 
54
63
  # Debug mode
55
64
  ai-speedometer --debug
65
+
66
+ # Headless mode - run benchmark directly
67
+ ai-speedometer --bench openai:gpt-4
68
+ # With custom API key
69
+ ai-speedometer --bench openai:gpt-4 --api-key "sk-your-key"
70
+ # Use AI SDK instead of REST API
71
+ ai-speedometer --bench openai:gpt-4 --ai-sdk
56
72
  ```
57
73
 
58
74
  ## Configuration Files