adaptive-memory-multi-model-router 2.4.0 β†’ 2.4.1

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.
Files changed (2) hide show
  1. package/README.md +49 -41
  2. package/package.json +1 -1
package/README.md CHANGED
@@ -1,17 +1,24 @@
1
1
  [πŸ‡¨πŸ‡³ δΈ­ζ–‡](./README_zh.md) Β· [πŸ‡―πŸ‡΅ ζ—₯本θͺž](./README_ja.md) Β· [English](./README.md)
2
2
 
3
- # A3M Router πŸ”€
3
+ # A3M Router
4
4
 
5
5
  [![npm](https://img.shields.io/npm/dt/adaptive-memory-multi-model-router?label=npm)](https://www.npmjs.com/package/adaptive-memory-multi-model-router)
6
6
  [![npm](https://img.shields.io/npm/v/adaptive-memory-multi-model-router)](https://www.npmjs.com/package/adaptive-memory-multi-model-router)
7
- [![GitHub stars](https://img.shields.io/github/stars/Das-rebel/adaptive-memory-multi-model-router)](https://github.com/Das-rebel/adaptive-memory-multi-model-router)
8
- [![License](https://img.shields.io/github/license/Das-rebel/adaptive-memory-multi-model-router)](LICENSE)
9
7
  [![Tests](https://img.shields.io/badge/tests-55%2F55%20passing-brightgreen)](test.js)
8
+ [![License](https://img.shields.io/github/license/Das-rebel/adaptive-memory-multi-model-router)](LICENSE)
9
+ [![GitHub stars](https://img.shields.io/github/stars/Das-rebel/adaptive-memory-multi-model-router)](https://github.com/Das-rebel/adaptive-memory-multi-model-router)
10
+ [![Discord](https://img.shields.io/badge/Discord-Join-brightgreen?logo=discord)](https://discord.gg/a3m-router)
11
+ [![Docs](https://img.shields.io/badge/Docs-API-blue?logo=read-the-docs)](docs/API.md)
12
+
13
+ [Documentation](docs/API.md) Β· [GitHub](https://github.com/Das-rebel/adaptive-memory-multi-model-router) Β· [npm](https://www.npmjs.com/package/adaptive-memory-multi-model-router)
14
+
15
+ ---
16
+
17
+ ## What is A3M Router?
18
+
19
+ **Intelligent LLM routing with adaptive memory** β€” routes every query to the cheapest capable model across 36+ providers. 99.5% Β±1 tier accuracy. Zero ML weights. Zero GPU. Starts in <100ms.
10
20
 
11
- > **Intelligent LLM routing** β€” 99.5% Β±1 tier accuracy, zero ML, zero GPU.
12
- >
13
- > OpenAI-compatible proxy. Routes to cheapest capable model across 36+ providers.
14
- > Adaptive memory learns from usage. Built-in cache, guardrails, cost analytics.
21
+ A3M Router is an OpenAI-compatible proxy that uses multi-signal heuristic routing to classify query complexity and automatically select the most cost-effective provider.
15
22
 
16
23
  ## Install
17
24
 
@@ -21,19 +28,6 @@ pip install a3m-router # Python
21
28
  npx a3m-router serve # Start proxy: localhost:8787
22
29
  ```
23
30
 
24
- ## How It Works
25
-
26
- ```
27
- Query β†’ Guardrails (<1ms) β†’ Cache (<1ms) β†’ Classify (2ms) β†’ Route
28
- β”‚
29
- β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”Όβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
30
- β–Ό β–Ό β–Ό
31
- FREE CHEAP PREMIUM
32
- Groq, etc. Llama, Mistral GPT-4o, Claude
33
- ```
34
-
35
- **Routing signals:** Domain (legal, medical, code) + Task type + Query structure β†’ Tier β†’ Cheapest available
36
-
37
31
  ## Quick Start
38
32
 
39
33
  ### TypeScript
@@ -68,9 +62,9 @@ async with A3MRouter() as router:
68
62
  ### CLI
69
63
 
70
64
  ```bash
71
- npx a3m-router route "Explain quantum computing" # Get routing decision
65
+ npx a3m-router route "Explain quantum computing" # Get routing decision
72
66
  npx a3m-router serve --port 8787 # Start proxy
73
- npx a3m-router benchmark # Run accuracy test
67
+ npx a3m-router benchmark # Run accuracy test
74
68
  ```
75
69
 
76
70
  ### REST API
@@ -81,16 +75,30 @@ curl http://localhost:8787/v1/chat/completions \
81
75
  -d '{"model": "auto", "messages": [{"role": "user", "content": "Hi"}]}'
82
76
  ```
83
77
 
78
+ ## How It Works
79
+
80
+ ```
81
+ Query β†’ Guardrails (<1ms) β†’ Cache (<1ms) β†’ Classify (2ms) β†’ Route
82
+ β”‚
83
+ β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”Όβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
84
+ β–Ό β–Ό β–Ό
85
+ FREE CHEAP PREMIUM
86
+ Groq, etc. Llama, Mistral GPT-4o, Claude
87
+ ```
88
+
89
+ **Routing signals:** Domain + Task type + Query structure β†’ Tier β†’ Cheapest available
90
+
84
91
  ## Features
85
92
 
86
- | Feature | Description |
87
- |---------|-------------|
88
- | **Semantic Cache** | Trigram Jaccard, 30% hit rate, no GPU |
89
- | **Guardrails** | 17-pattern injection detection, PII redaction |
90
- | **Adaptive Memory** | EMA quality scoring, learns over time |
91
- | **Circuit Breaker** | 3 failures β†’ 60s cooldown, auto failover |
92
- | **Cost Analytics** | Per-provider spend, budget alerts |
93
- | **OpenAI-Compatible** | Drop-in for any OpenAI SDK |
93
+ | | |
94
+ |---|---|
95
+ | **Adaptive Memory** | Learns from usage. EMA quality scoring. No retraining. |
96
+ | **Multi-Signal Routing** | 12 signals across 5 dimensions. Zero ML weights. |
97
+ | **Semantic Cache** | Trigram Jaccard similarity. 30% hit rate. |
98
+ | **Guardrails** | 17-pattern injection detection. PII redaction. |
99
+ | **Circuit Breaker** | 3 failures β†’ 60s cooldown. Auto failover. |
100
+ | **Cost Analytics** | Per-provider spend. Budget alerts. Savings tracking. |
101
+ | **OpenAI-Compatible** | Drop-in for any OpenAI SDK. |
94
102
 
95
103
  ## Benchmark
96
104
 
@@ -109,21 +117,21 @@ Run: `node scripts/routing-benchmark-v2.js`
109
117
 
110
118
  | Tier | Score | Examples |
111
119
  |------|-------|----------|
112
- | **FREE** | <0.20 | Groq (LLaMA 3.3 70B), DeepSeek Chat |
113
- | **CHEAP** | 0.20-0.45 | Mistral 7B, Qwen 2.5, Yi Large |
114
- | **MID** | 0.45-0.65 | Claude 3 Haiku, GPT-4o-mini |
115
- | **PREMIUM** | >0.65 | Claude 3.5 Sonnet, GPT-4o |
120
+ | **FREE** | <0.20 | Groq (LLaMA 3.3 70B), DeepSeek Chat, Fireworks |
121
+ | **CHEAP** | 0.20-0.45 | Mistral 7B, Qwen 2.5, Yi Large, DeepSeek Coder |
122
+ | **MID** | 0.45-0.65 | Claude 3 Haiku, GPT-4o-mini, Gemini Pro |
123
+ | **PREMIUM** | >0.65 | Claude 3.5 Sonnet, GPT-4o, Gemini Ultra |
116
124
 
117
- **Chinese:** Kimi (Moonshot), Zhipu GLM, Qwen, Yi, Baichuan, StepFun
125
+ **Chinese providers:** Kimi (Moonshot), Zhipu GLM, Qwen, Yi, Baichuan, StepFun
118
126
 
119
127
  ## Comparison
120
128
 
121
129
  | | A3M Router | LiteLLM | Portkey |
122
130
  |--|:--:|:--:|:--:|
123
131
  | Auto Routing | βœ“ | Manual | Manual |
124
- | Semantic Cache | βœ“ 30% | - | Limited |
125
- | Guardrails | βœ“ 17 patterns | - | Limited |
126
- | Circuit Breaker | βœ“ | - | - |
132
+ | Semantic Cache | βœ“ 30% | β€” | Limited |
133
+ | Guardrails | βœ“ 17 patterns | β€” | Limited |
134
+ | Circuit Breaker | βœ“ | β€” | β€” |
127
135
  | Package Size | 19.5 KB | ~50 MB | ~30 MB |
128
136
  | Startup | <100ms | ~500ms | ~300ms |
129
137
 
@@ -136,9 +144,9 @@ Run: `node scripts/routing-benchmark-v2.js`
136
144
 
137
145
  ## Links
138
146
 
139
- - [npm](https://www.npmjs.com/package/adaptive-memory-multi-model-router)
140
- - [GitHub](https://github.com/Das-rebel/adaptive-memory-multi-model-router)
141
- - [API docs](docs/API.md)
147
+ - [Documentation](docs/API.md)
148
+ - [npm package](https://www.npmjs.com/package/adaptive-memory-multi-model-router)
149
+ - [GitHub repo](https://github.com/Das-rebel/adaptive-memory-multi-model-router)
142
150
  - [Contributing](CONTRIBUTING.md)
143
151
 
144
152
  MIT License. `npm install` and go.
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "adaptive-memory-multi-model-router",
3
- "version": "2.4.0",
3
+ "version": "2.4.1",
4
4
  "shortName": "A3M Router",
5
5
  "displayName": "A3M Router - Adaptive Memory Multi-Model Router",
6
6
  "description": "LLM router & AI gateway with 99.5% routing accuracy β€” supports 47 providers including DeepSeek, Kimi (Moonshot), Qwen, Zhipu GLM, Yi, Baichuan, MiniMax, StepFun. Zero ML, 19.5KB. Multi-signal routing, semantic cache, guardrails, cost analytics. MIT. TypeScript SDK + Python SDK + OpenAI proxy.",