adaptive-memory-multi-model-router 2.4.1 → 2.5.0
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 +764 -82
- package/package.json +410 -94
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "adaptive-memory-multi-model-router",
|
|
3
|
-
"version": "2.
|
|
3
|
+
"version": "2.5.0",
|
|
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 — 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.",
|
|
@@ -53,118 +53,434 @@
|
|
|
53
53
|
}
|
|
54
54
|
},
|
|
55
55
|
"keywords": [
|
|
56
|
-
"
|
|
57
|
-
"
|
|
58
|
-
"
|
|
59
|
-
"
|
|
56
|
+
"adaptive-router",
|
|
57
|
+
"agent",
|
|
58
|
+
"agent-framework",
|
|
59
|
+
"ai-agent-tools",
|
|
60
|
+
"ai-assistant",
|
|
61
|
+
"ai-discoverability",
|
|
62
|
+
"ai-routing",
|
|
63
|
+
"airtable",
|
|
64
|
+
"amplitude",
|
|
65
|
+
"api-gateway",
|
|
66
|
+
"asana",
|
|
67
|
+
"automation",
|
|
68
|
+
"autonomous-agents",
|
|
69
|
+
"aws-bedrock",
|
|
70
|
+
"batch-processing",
|
|
71
|
+
"bitbucket",
|
|
72
|
+
"circuit-breaker",
|
|
73
|
+
"claude-code",
|
|
74
|
+
"cli",
|
|
75
|
+
"cloud",
|
|
76
|
+
"code-generation",
|
|
77
|
+
"cohere",
|
|
78
|
+
"content-filtering",
|
|
79
|
+
"copilot",
|
|
80
|
+
"cost-analytics",
|
|
81
|
+
"datadog",
|
|
82
|
+
"deepinfra",
|
|
83
|
+
"developer-experience",
|
|
84
|
+
"developer-tools",
|
|
85
|
+
"devops",
|
|
86
|
+
"discord",
|
|
87
|
+
"dropbox",
|
|
88
|
+
"dx",
|
|
89
|
+
"edge-computing",
|
|
90
|
+
"embedding",
|
|
91
|
+
"enterprise",
|
|
92
|
+
"fallback",
|
|
93
|
+
"fine-tuning",
|
|
94
|
+
"fireworks",
|
|
95
|
+
"gemini",
|
|
96
|
+
"generative-engine-optimization",
|
|
97
|
+
"geo",
|
|
98
|
+
"github",
|
|
99
|
+
"gitlab",
|
|
100
|
+
"gmail",
|
|
101
|
+
"google",
|
|
102
|
+
"google-calendar",
|
|
103
|
+
"gpt-4",
|
|
104
|
+
"graphql",
|
|
105
|
+
"high-availability",
|
|
106
|
+
"hubspot",
|
|
107
|
+
"huggingface",
|
|
108
|
+
"inference",
|
|
109
|
+
"input-validation",
|
|
110
|
+
"intent-mapping",
|
|
111
|
+
"intercom",
|
|
112
|
+
"javascript",
|
|
113
|
+
"jira",
|
|
114
|
+
"latency-optimization",
|
|
115
|
+
"linear",
|
|
116
|
+
"llamaindex",
|
|
117
|
+
"llm-intent",
|
|
118
|
+
"llm-tools",
|
|
119
|
+
"llmlingua",
|
|
120
|
+
"load-balancing",
|
|
121
|
+
"logging",
|
|
122
|
+
"machine-learning",
|
|
123
|
+
"mailchimp",
|
|
124
|
+
"mcp",
|
|
125
|
+
"medusa",
|
|
126
|
+
"memory-tree",
|
|
127
|
+
"middleware",
|
|
128
|
+
"mixpanel",
|
|
129
|
+
"monitoring",
|
|
130
|
+
"multi-model",
|
|
131
|
+
"natural-language-processing",
|
|
132
|
+
"netlify",
|
|
133
|
+
"nodejs",
|
|
134
|
+
"notion",
|
|
135
|
+
"npm",
|
|
136
|
+
"observability",
|
|
137
|
+
"open-source",
|
|
138
|
+
"orchestration",
|
|
139
|
+
"parallel-execution",
|
|
140
|
+
"performance",
|
|
141
|
+
"perplexity",
|
|
142
|
+
"pii-detection",
|
|
143
|
+
"pinecone",
|
|
144
|
+
"posthog",
|
|
145
|
+
"prefix-caching",
|
|
146
|
+
"production",
|
|
147
|
+
"prompt-engineering",
|
|
148
|
+
"prompt-injection",
|
|
149
|
+
"provider-registry",
|
|
150
|
+
"proxy",
|
|
151
|
+
"proxy-server",
|
|
152
|
+
"python",
|
|
153
|
+
"radix-attention",
|
|
154
|
+
"rag",
|
|
155
|
+
"rate-limiting",
|
|
156
|
+
"real-time",
|
|
157
|
+
"rest-api",
|
|
158
|
+
"retrieval-augmented-generation",
|
|
159
|
+
"retry",
|
|
160
|
+
"route-quality",
|
|
161
|
+
"routellm",
|
|
162
|
+
"router",
|
|
163
|
+
"s3",
|
|
164
|
+
"salesforce",
|
|
165
|
+
"sanitization",
|
|
166
|
+
"scalability",
|
|
167
|
+
"sdk",
|
|
168
|
+
"security",
|
|
169
|
+
"segment",
|
|
170
|
+
"sendgrid",
|
|
171
|
+
"sentry",
|
|
172
|
+
"serverless",
|
|
173
|
+
"shopify",
|
|
174
|
+
"slack",
|
|
175
|
+
"speculative-decoding",
|
|
176
|
+
"streaming",
|
|
177
|
+
"stripe",
|
|
178
|
+
"telegram",
|
|
179
|
+
"testing",
|
|
180
|
+
"together-ai",
|
|
181
|
+
"token-compression",
|
|
182
|
+
"tools",
|
|
183
|
+
"tracing",
|
|
184
|
+
"transformer",
|
|
185
|
+
"trello",
|
|
186
|
+
"typescript",
|
|
187
|
+
"vector-database",
|
|
188
|
+
"vercel",
|
|
189
|
+
"websocket",
|
|
190
|
+
"xai",
|
|
191
|
+
"zendesk",
|
|
192
|
+
"llm-proxy",
|
|
60
193
|
"claude",
|
|
61
|
-
"
|
|
194
|
+
"ai",
|
|
195
|
+
"openai-compatible",
|
|
196
|
+
"no-gpu",
|
|
197
|
+
"openai",
|
|
198
|
+
"routing-accuracy",
|
|
199
|
+
"minimax",
|
|
200
|
+
"portkey-alternative",
|
|
201
|
+
"中文llm",
|
|
202
|
+
"provider-fallback",
|
|
203
|
+
"nlp",
|
|
204
|
+
"ollama",
|
|
205
|
+
"cerebras",
|
|
206
|
+
"openai-sdk",
|
|
62
207
|
"ai-sdk",
|
|
63
|
-
"qwen",
|
|
64
|
-
"groq",
|
|
65
|
-
"high-availability",
|
|
66
208
|
"openai-proxy",
|
|
67
|
-
"
|
|
209
|
+
"cerebras-proxy",
|
|
210
|
+
"深度求索",
|
|
211
|
+
"anthropic",
|
|
212
|
+
"ai-cost-optimization",
|
|
68
213
|
"model-router",
|
|
69
|
-
"
|
|
70
|
-
"
|
|
71
|
-
"semantic-routing",
|
|
72
|
-
"circuit-breaker",
|
|
73
|
-
"openrouter",
|
|
74
|
-
"litellm-alternative",
|
|
75
|
-
"zhipu",
|
|
76
|
-
"llm-management",
|
|
77
|
-
"ai-agent",
|
|
214
|
+
"model-routing",
|
|
215
|
+
"chinese-ai",
|
|
78
216
|
"vercel-ai",
|
|
79
|
-
"
|
|
80
|
-
"
|
|
217
|
+
"ai-load-balancer",
|
|
218
|
+
"deepseek",
|
|
219
|
+
"llm-routing",
|
|
81
220
|
"mistral",
|
|
82
|
-
"
|
|
83
|
-
"
|
|
84
|
-
"ollama",
|
|
85
|
-
"cost-optimization",
|
|
86
|
-
"chatgpt",
|
|
87
|
-
"nlp",
|
|
88
|
-
"smart-llm-router",
|
|
89
|
-
"openai-sdk",
|
|
90
|
-
"distributed-llm",
|
|
91
|
-
"llm-proxy",
|
|
92
|
-
"chatbot",
|
|
93
|
-
"ai-cost-optimization",
|
|
221
|
+
"ollama-proxy",
|
|
222
|
+
"moonshot",
|
|
94
223
|
"stepfun",
|
|
95
|
-
"
|
|
96
|
-
"
|
|
97
|
-
"
|
|
98
|
-
"
|
|
99
|
-
"
|
|
100
|
-
"
|
|
101
|
-
"
|
|
102
|
-
"
|
|
103
|
-
"ai",
|
|
104
|
-
"intent-routing",
|
|
105
|
-
"api-gateway",
|
|
106
|
-
"retry-proxy",
|
|
224
|
+
"cost-optimization",
|
|
225
|
+
"keyword-routing",
|
|
226
|
+
"zhipu",
|
|
227
|
+
"llm-manager",
|
|
228
|
+
"ai-router",
|
|
229
|
+
"multi-provider",
|
|
230
|
+
"yi-llm",
|
|
231
|
+
"chinese-llm",
|
|
107
232
|
"glms",
|
|
233
|
+
"llm",
|
|
234
|
+
"llm-load-balancer",
|
|
235
|
+
"language-model",
|
|
108
236
|
"claude-proxy",
|
|
109
|
-
"content-routing",
|
|
110
|
-
"chinese-llm",
|
|
111
237
|
"lightweight",
|
|
112
|
-
"
|
|
113
|
-
"
|
|
114
|
-
"ai-orchestration",
|
|
115
|
-
"provider-fallback",
|
|
116
|
-
"zh-llm",
|
|
117
|
-
"load-balancer",
|
|
118
|
-
"request-routing",
|
|
119
|
-
"model-routing",
|
|
120
|
-
"anthropic",
|
|
238
|
+
"chatgpt",
|
|
239
|
+
"smart-routing",
|
|
121
240
|
"阿里qwen",
|
|
122
|
-
"
|
|
123
|
-
"深度求索",
|
|
124
|
-
"intelligent-router",
|
|
125
|
-
"domain-routing",
|
|
241
|
+
"zh-llm",
|
|
126
242
|
"中文ai",
|
|
127
|
-
"aleph-alpha",
|
|
128
|
-
"llm-router",
|
|
129
|
-
"minimax",
|
|
130
243
|
"benchmark",
|
|
131
|
-
"
|
|
132
|
-
"
|
|
133
|
-
"
|
|
134
|
-
"
|
|
135
|
-
"
|
|
136
|
-
"
|
|
244
|
+
"helicone",
|
|
245
|
+
"mistral-proxy",
|
|
246
|
+
"llm-orchestration",
|
|
247
|
+
"openrouter",
|
|
248
|
+
"kimi",
|
|
249
|
+
"moonshot-kimi",
|
|
250
|
+
"guardrails",
|
|
251
|
+
"ai-agent",
|
|
137
252
|
"baichuan",
|
|
138
|
-
"
|
|
139
|
-
"openai",
|
|
140
|
-
"language-model",
|
|
141
|
-
"ai-router",
|
|
142
|
-
"yi-llm",
|
|
253
|
+
"routellm-alternative",
|
|
143
254
|
"anthropic-proxy",
|
|
144
|
-
"
|
|
255
|
+
"token-counter",
|
|
256
|
+
"deepseek-proxy",
|
|
257
|
+
"llm-gateway",
|
|
258
|
+
"multi-llm",
|
|
259
|
+
"ai-gateway",
|
|
260
|
+
"groq",
|
|
261
|
+
"litellm-alternative",
|
|
262
|
+
"ai-proxy",
|
|
263
|
+
"llm-router",
|
|
264
|
+
"domain-routing",
|
|
265
|
+
"groq-proxy",
|
|
266
|
+
"qwen",
|
|
145
267
|
"gpt",
|
|
146
|
-
"chinese-ai",
|
|
147
|
-
"moonshot-kimi",
|
|
148
|
-
"routing-accuracy",
|
|
149
|
-
"cerebras-proxy",
|
|
150
268
|
"openai-api",
|
|
151
|
-
"
|
|
152
|
-
"
|
|
153
|
-
"
|
|
154
|
-
"ai-load-balancer",
|
|
155
|
-
"routellm-alternative",
|
|
156
|
-
"token-counter",
|
|
157
|
-
"smart-routing",
|
|
158
|
-
"model-management",
|
|
159
|
-
"deepseek",
|
|
160
|
-
"task-routing",
|
|
161
|
-
"ollama-proxy",
|
|
269
|
+
"query-routing",
|
|
270
|
+
"aleph-alpha",
|
|
271
|
+
"semantic-cache",
|
|
162
272
|
"langchain",
|
|
273
|
+
"ai-guardrails",
|
|
274
|
+
"chatbot",
|
|
275
|
+
"a3m",
|
|
276
|
+
"a3m-router",
|
|
277
|
+
"adaptive",
|
|
278
|
+
"memory-based",
|
|
279
|
+
"multi-model-router",
|
|
280
|
+
"memory-based-router",
|
|
281
|
+
"treequest",
|
|
282
|
+
"parallel-ai",
|
|
283
|
+
"agent-orchestration",
|
|
284
|
+
"multi-agent",
|
|
285
|
+
"parallel",
|
|
286
|
+
"cost-tracking",
|
|
287
|
+
"cache",
|
|
288
|
+
"caching",
|
|
289
|
+
"exponential-backoff",
|
|
290
|
+
"mcts",
|
|
291
|
+
"monte-carlo-tree-search",
|
|
292
|
+
"workflow-optimization",
|
|
293
|
+
"hierarchical-planning",
|
|
294
|
+
"halo",
|
|
295
|
+
"episodic-memory",
|
|
296
|
+
"semantic-memory",
|
|
297
|
+
"agent-memory",
|
|
298
|
+
"python-bindings",
|
|
299
|
+
"pypi",
|
|
300
|
+
"autogen",
|
|
301
|
+
"crewai",
|
|
302
|
+
"transformers",
|
|
303
|
+
"agent-codegen",
|
|
304
|
+
"ai-coding",
|
|
305
|
+
"zai",
|
|
306
|
+
"llama",
|
|
307
|
+
"ai-agents",
|
|
308
|
+
"memory-based-llm-router",
|
|
309
|
+
"multi-llm-router",
|
|
310
|
+
"llm-memory-router",
|
|
311
|
+
"adaptive-llm-router",
|
|
312
|
+
"intelligent-router",
|
|
313
|
+
"intelligent-llm-router",
|
|
314
|
+
"learning-router",
|
|
315
|
+
"contextual-router",
|
|
316
|
+
"context-aware-router",
|
|
317
|
+
"task-aware-router",
|
|
318
|
+
"memory-augmented",
|
|
319
|
+
"memory-augmented-llm",
|
|
320
|
+
"episodic-memory-router",
|
|
321
|
+
"semantic-memory-router",
|
|
322
|
+
"task-memory",
|
|
323
|
+
"cross-context-memory",
|
|
324
|
+
"context-compression",
|
|
325
|
+
"ison-format",
|
|
326
|
+
"message-truncation",
|
|
327
|
+
"context-management",
|
|
328
|
+
"local-llm",
|
|
329
|
+
"lmstudio",
|
|
330
|
+
"local-model",
|
|
331
|
+
"privacy-llm",
|
|
332
|
+
"priority-queue",
|
|
333
|
+
"token-counting",
|
|
334
|
+
"cost-estimation",
|
|
335
|
+
"cost-prediction",
|
|
336
|
+
"intelligent-failover",
|
|
337
|
+
"kv-cache",
|
|
338
|
+
"pagedattention",
|
|
339
|
+
"kv-cache-quantization",
|
|
340
|
+
"streamingllm",
|
|
341
|
+
"multimodel-orchestration",
|
|
342
|
+
"multi-agent-debate",
|
|
343
|
+
"self-consistency",
|
|
344
|
+
"tensor-parallelism",
|
|
345
|
+
"continuous-batching",
|
|
346
|
+
"arxiv",
|
|
347
|
+
"research-backed",
|
|
348
|
+
"icml",
|
|
349
|
+
"neurips",
|
|
350
|
+
"iclr",
|
|
351
|
+
"pi-extension",
|
|
352
|
+
"pi",
|
|
353
|
+
"pi-package",
|
|
354
|
+
"pi-coding-agent",
|
|
355
|
+
"pi-agent",
|
|
356
|
+
"agent-discoverable",
|
|
357
|
+
"ai-native",
|
|
358
|
+
"01-ai",
|
|
359
|
+
"01ai",
|
|
360
|
+
"128k上下文",
|
|
361
|
+
"ai-gateway-cn",
|
|
362
|
+
"ai路由",
|
|
363
|
+
"aliyun",
|
|
364
|
+
"baichuan-ai",
|
|
365
|
+
"baichuan2",
|
|
366
|
+
"baichuan2-flash",
|
|
367
|
+
"baiducloud",
|
|
368
|
+
"chinese-ai-proxy",
|
|
369
|
+
"chinese-api-gateway",
|
|
370
|
+
"chinese-chatgpt",
|
|
371
|
+
"chinese-language-model",
|
|
372
|
+
"chinese-llm-proxy",
|
|
373
|
+
"chinese-models",
|
|
374
|
+
"chinese-nlp",
|
|
375
|
+
"chinesellm",
|
|
376
|
+
"chinesellm-routing",
|
|
377
|
+
"deepseek-ai",
|
|
378
|
+
"deepseek-api",
|
|
379
|
+
"deepseek-chat",
|
|
380
|
+
"deepseek-coder",
|
|
381
|
+
"deepseek-reasoner",
|
|
382
|
+
"embedding-model",
|
|
383
|
+
"glm-4",
|
|
384
|
+
"glm-4-air",
|
|
385
|
+
"glm-4-flash",
|
|
386
|
+
"glm-4-long",
|
|
387
|
+
"glm-4-plus",
|
|
388
|
+
"huawei-cloud",
|
|
389
|
+
"langchain-cn",
|
|
390
|
+
"lingyi",
|
|
391
|
+
"lingyi-wanwu",
|
|
392
|
+
"llamaindex-cn",
|
|
393
|
+
"llm-gateway-cn",
|
|
394
|
+
"long-context-model",
|
|
395
|
+
"minimax-ai",
|
|
396
|
+
"minimax-api",
|
|
397
|
+
"minimax-chat",
|
|
398
|
+
"minimax-turbo",
|
|
399
|
+
"moonshot-ai",
|
|
400
|
+
"moonshot-v1",
|
|
401
|
+
"moonshot-v1-128k",
|
|
402
|
+
"moonshot-v1-32k",
|
|
403
|
+
"moonshot-v1-8k",
|
|
404
|
+
"multimodal",
|
|
405
|
+
"qianwen",
|
|
406
|
+
"qwen-long",
|
|
407
|
+
"qwen-max",
|
|
408
|
+
"qwen-plus",
|
|
409
|
+
"qwen-turbo",
|
|
410
|
+
"qwen2",
|
|
411
|
+
"qwen3",
|
|
412
|
+
"step-1",
|
|
413
|
+
"step-1v",
|
|
414
|
+
"step-2",
|
|
415
|
+
"tencentcloud",
|
|
416
|
+
"tongji",
|
|
417
|
+
"tongyi",
|
|
418
|
+
"tongyi-qianwen",
|
|
419
|
+
"vision-model",
|
|
420
|
+
"volcengine",
|
|
421
|
+
"wu-yuan",
|
|
422
|
+
"wuyuan",
|
|
423
|
+
"yi-ai",
|
|
424
|
+
"yi-large",
|
|
425
|
+
"yi-lightning",
|
|
426
|
+
"yi-medium",
|
|
427
|
+
"zai-glm",
|
|
428
|
+
"zai-glm-4",
|
|
429
|
+
"zhipu-ai",
|
|
430
|
+
"中国ai",
|
|
431
|
+
"中国大模型",
|
|
432
|
+
"中文chatgpt",
|
|
433
|
+
"中文embedding",
|
|
434
|
+
"中文langchain",
|
|
435
|
+
"中文llamaindex",
|
|
436
|
+
"中文nlp",
|
|
437
|
+
"中文seo",
|
|
438
|
+
"中文多模态",
|
|
439
|
+
"中文大模型",
|
|
440
|
+
"中文搜索引擎优化",
|
|
441
|
+
"中文模型",
|
|
442
|
+
"华为云",
|
|
443
|
+
"向量化",
|
|
444
|
+
"国产ai",
|
|
445
|
+
"国产大模型",
|
|
446
|
+
"多模型路由",
|
|
447
|
+
"多模态",
|
|
448
|
+
"大模型路由",
|
|
449
|
+
"字节ai",
|
|
450
|
+
"智能路由",
|
|
451
|
+
"火山引擎",
|
|
452
|
+
"百度ai",
|
|
453
|
+
"百度云",
|
|
454
|
+
"腾讯ai",
|
|
455
|
+
"腾讯云",
|
|
456
|
+
"长上下文",
|
|
457
|
+
"阿里ai",
|
|
458
|
+
"阿里云",
|
|
459
|
+
"adaptive-routing",
|
|
460
|
+
"ai-orchestration",
|
|
461
|
+
"content-routing",
|
|
462
|
+
"cost-based-routing",
|
|
463
|
+
"distributed-llm",
|
|
464
|
+
"dynamic-routing",
|
|
465
|
+
"failover",
|
|
466
|
+
"intelligent-load-balancer",
|
|
467
|
+
"intent-routing",
|
|
468
|
+
"llm-caching",
|
|
469
|
+
"llm-cluster",
|
|
163
470
|
"llm-failover",
|
|
164
|
-
"
|
|
165
|
-
"
|
|
166
|
-
"
|
|
167
|
-
"
|
|
471
|
+
"llm-management",
|
|
472
|
+
"load-balancer",
|
|
473
|
+
"model-management",
|
|
474
|
+
"model-selection",
|
|
475
|
+
"provider-selector",
|
|
476
|
+
"rate-limiter",
|
|
477
|
+
"request-routing",
|
|
478
|
+
"retry-proxy",
|
|
479
|
+
"reverse-proxy",
|
|
480
|
+
"semantic-routing",
|
|
481
|
+
"smart-llm-router",
|
|
482
|
+
"task-routing",
|
|
483
|
+
"traffic-routing"
|
|
168
484
|
],
|
|
169
485
|
"author": "Das-rebel <subho@example.com>",
|
|
170
486
|
"license": "MIT",
|