adaptive-memory-multi-model-router 2.2.6 → 2.2.8

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "adaptive-memory-multi-model-router",
-  "version": "2.2.6",
+  "version": "2.2.8",
   "shortName": "A3M Router",
   "displayName": "A3M Router - Adaptive Memory Multi-Model Router",
   "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.",
@@ -53,409 +53,118 @@
     }
   },
   "keywords": [
-    "adaptive-router",
-    "agent",
-    "agent-framework",
-    "ai-agent-tools",
-    "ai-assistant",
-    "ai-discoverability",
-    "ai-routing",
-    "airtable",
-    "amplitude",
-    "api-gateway",
-    "asana",
-    "automation",
-    "autonomous-agents",
-    "aws-bedrock",
-    "batch-processing",
-    "bitbucket",
-    "circuit-breaker",
-    "claude-code",
-    "cli",
-    "cloud",
-    "code-generation",
-    "cohere",
-    "content-filtering",
-    "copilot",
-    "cost-analytics",
-    "datadog",
-    "deepinfra",
-    "developer-experience",
-    "developer-tools",
-    "devops",
-    "discord",
-    "dropbox",
-    "dx",
-    "edge-computing",
-    "embedding",
-    "enterprise",
-    "fallback",
-    "fine-tuning",
-    "fireworks",
-    "gemini",
-    "generative-engine-optimization",
-    "geo",
-    "github",
-    "gitlab",
-    "gmail",
-    "google",
-    "google-calendar",
-    "gpt-4",
-    "graphql",
-    "high-availability",
-    "hubspot",
-    "huggingface",
-    "inference",
-    "input-validation",
-    "intent-mapping",
-    "intercom",
-    "javascript",
-    "jira",
-    "latency-optimization",
-    "linear",
-    "llamaindex",
-    "llm-intent",
-    "llm-tools",
-    "llmlingua",
-    "load-balancing",
-    "logging",
-    "machine-learning",
-    "mailchimp",
-    "mcp",
-    "medusa",
-    "memory-tree",
-    "middleware",
-    "mixpanel",
-    "monitoring",
-    "multi-model",
-    "natural-language-processing",
-    "netlify",
-    "nodejs",
-    "notion",
-    "npm",
-    "observability",
-    "open-source",
-    "orchestration",
-    "parallel-execution",
-    "performance",
-    "perplexity",
-    "pii-detection",
-    "pinecone",
-    "posthog",
-    "prefix-caching",
-    "production",
-    "prompt-engineering",
-    "prompt-injection",
-    "provider-registry",
-    "proxy",
-    "proxy-server",
-    "python",
-    "radix-attention",
-    "rag",
-    "rate-limiting",
-    "real-time",
-    "rest-api",
-    "retrieval-augmented-generation",
-    "retry",
-    "route-quality",
-    "routellm",
-    "router",
-    "s3",
-    "salesforce",
-    "sanitization",
-    "scalability",
-    "sdk",
-    "security",
-    "segment",
-    "sendgrid",
-    "sentry",
-    "serverless",
-    "shopify",
-    "slack",
-    "speculative-decoding",
-    "streaming",
-    "stripe",
-    "telegram",
-    "testing",
-    "together-ai",
-    "token-compression",
-    "tools",
-    "tracing",
-    "transformer",
-    "trello",
-    "typescript",
-    "vector-database",
-    "vercel",
-    "websocket",
-    "xai",
-    "zendesk",
-    "llm-proxy",
+    "ai-guardrails",
+    "moonshot",
+    "adaptive-routing",
+    "llm-manager",
     "claude",
-    "ai",
-    "openai-compatible",
-    "no-gpu",
-    "openai",
-    "routing-accuracy",
-    "minimax",
-    "portkey-alternative",
-    "中文llm",
-    "provider-fallback",
-    "nlp",
-    "ollama",
-    "cerebras",
-    "openai-sdk",
+    "guardrails",
     "ai-sdk",
+    "qwen",
+    "groq",
+    "high-availability",
     "openai-proxy",
-    "cerebras-proxy",
-    "深度求索",
-    "anthropic",
-    "ai-cost-optimization",
+    "multi-provider",
     "model-router",
-    "model-routing",
-    "chinese-ai",
-    "vercel-ai",
-    "ai-load-balancer",
-    "deepseek",
+    "rate-limiter",
     "llm-routing",
+    "semantic-routing",
+    "circuit-breaker",
+    "openrouter",
+    "litellm-alternative",
+    "zhipu",
+    "llm-management",
+    "ai-agent",
+    "vercel-ai",
+    "deepseek-proxy",
+    "groq-proxy",
     "mistral",
-    "ollama-proxy",
-    "moonshot",
-    "stepfun",
+    "llm-gateway",
+    "cost-based-routing",
+    "ollama",
     "cost-optimization",
-    "keyword-routing",
-    "zhipu",
-    "llm-manager",
-    "ai-router",
-    "multi-provider",
-    "yi-llm",
-    "chinese-llm",
-    "glms",
-    "llm",
+    "chatgpt",
+    "nlp",
+    "smart-llm-router",
+    "openai-sdk",
+    "distributed-llm",
+    "llm-proxy",
+    "chatbot",
+    "ai-cost-optimization",
+    "stepfun",
+    "reverse-proxy",
+    "llm-orchestration",
+    "llm-caching",
+    "intelligent-load-balancer",
     "llm-load-balancer",
-    "language-model",
+    "ai-gateway",
+    "helicone",
+    "failover",
+    "ai",
+    "intent-routing",
+    "api-gateway",
+    "retry-proxy",
+    "glms",
     "claude-proxy",
+    "content-routing",
+    "chinese-llm",
     "lightweight",
-    "chatgpt",
-    "smart-routing",
-    "阿里qwen",
+    "dynamic-routing",
+    "provider-selector",
+    "ai-orchestration",
+    "provider-fallback",
     "zh-llm",
+    "load-balancer",
+    "request-routing",
+    "model-routing",
+    "anthropic",
+    "阿里qwen",
+    "multi-llm",
+    "深度求索",
+    "intelligent-router",
+    "domain-routing",
     "中文ai",
+    "aleph-alpha",
+    "llm-router",
+    "minimax",
     "benchmark",
-    "helicone",
-    "mistral-proxy",
-    "llm-orchestration",
-    "openrouter",
-    "kimi",
-    "moonshot-kimi",
-    "guardrails",
-    "ai-agent",
+    "cerebras",
+    "keyword-routing",
+    "traffic-routing",
+    "中文llm",
+    "ai-proxy",
+    "llm-cluster",
     "baichuan",
-    "routellm-alternative",
+    "portkey-alternative",
+    "openai",
+    "language-model",
+    "ai-router",
+    "yi-llm",
     "anthropic-proxy",
-    "token-counter",
-    "deepseek-proxy",
-    "llm-gateway",
-    "multi-llm",
-    "ai-gateway",
-    "groq",
-    "litellm-alternative",
-    "ai-proxy",
-    "llm-router",
-    "domain-routing",
-    "groq-proxy",
-    "qwen",
+    "query-routing",
     "gpt",
+    "chinese-ai",
+    "moonshot-kimi",
+    "routing-accuracy",
+    "cerebras-proxy",
     "openai-api",
-    "query-routing",
-    "aleph-alpha",
-    "semantic-cache",
+    "kimi",
+    "model-selection",
+    "no-gpu",
+    "ai-load-balancer",
+    "routellm-alternative",
+    "token-counter",
+    "smart-routing",
+    "model-management",
+    "deepseek",
+    "task-routing",
+    "ollama-proxy",
     "langchain",
-    "ai-guardrails",
-    "chatbot",
-    "a3m",
-    "a3m-router",
-    "adaptive",
-    "memory-based",
-    "multi-model-router",
-    "memory-based-router",
-    "treequest",
-    "parallel-ai",
-    "agent-orchestration",
-    "multi-agent",
-    "parallel",
-    "cost-tracking",
-    "cache",
-    "caching",
-    "exponential-backoff",
-    "mcts",
-    "monte-carlo-tree-search",
-    "workflow-optimization",
-    "hierarchical-planning",
-    "halo",
-    "episodic-memory",
-    "semantic-memory",
-    "agent-memory",
-    "python-bindings",
-    "pypi",
-    "autogen",
-    "crewai",
-    "transformers",
-    "agent-codegen",
-    "ai-coding",
-    "zai",
-    "llama",
-    "ai-agents",
-    "memory-based-llm-router",
-    "multi-llm-router",
-    "llm-memory-router",
-    "adaptive-llm-router",
-    "intelligent-router",
-    "intelligent-llm-router",
-    "learning-router",
-    "contextual-router",
-    "context-aware-router",
-    "task-aware-router",
-    "memory-augmented",
-    "memory-augmented-llm",
-    "episodic-memory-router",
-    "semantic-memory-router",
-    "task-memory",
-    "cross-context-memory",
-    "context-compression",
-    "ison-format",
-    "message-truncation",
-    "context-management",
-    "local-llm",
-    "lmstudio",
-    "local-model",
-    "privacy-llm",
-    "priority-queue",
-    "token-counting",
-    "cost-estimation",
-    "cost-prediction",
-    "intelligent-failover",
-    "kv-cache",
-    "pagedattention",
-    "kv-cache-quantization",
-    "streamingllm",
-    "multimodel-orchestration",
-    "multi-agent-debate",
-    "self-consistency",
-    "tensor-parallelism",
-    "continuous-batching",
-    "arxiv",
-    "research-backed",
-    "icml",
-    "neurips",
-    "iclr",
-    "pi-extension",
-    "pi",
-    "pi-package",
-    "pi-coding-agent",
-    "pi-agent",
-    "agent-discoverable",
-    "ai-native",
-    "01-ai",
-    "01ai",
-    "128k上下文",
-    "ai-gateway-cn",
-    "ai路由",
-    "aliyun",
-    "baichuan-ai",
-    "baichuan2",
-    "baichuan2-flash",
-    "baiducloud",
-    "chinese-ai-proxy",
-    "chinese-api-gateway",
-    "chinese-chatgpt",
-    "chinese-language-model",
-    "chinese-llm-proxy",
-    "chinese-models",
-    "chinese-nlp",
-    "chinesellm",
-    "chinesellm-routing",
-    "deepseek-ai",
-    "deepseek-api",
-    "deepseek-chat",
-    "deepseek-coder",
-    "deepseek-reasoner",
-    "embedding-model",
-    "glm-4",
-    "glm-4-air",
-    "glm-4-flash",
-    "glm-4-long",
-    "glm-4-plus",
-    "huawei-cloud",
-    "langchain-cn",
-    "lingyi",
-    "lingyi-wanwu",
-    "llamaindex-cn",
-    "llm-gateway-cn",
-    "long-context-model",
-    "minimax-ai",
-    "minimax-api",
-    "minimax-chat",
-    "minimax-turbo",
-    "moonshot-ai",
-    "moonshot-v1",
-    "moonshot-v1-128k",
-    "moonshot-v1-32k",
-    "moonshot-v1-8k",
-    "multimodal",
-    "qianwen",
-    "qwen-long",
-    "qwen-max",
-    "qwen-plus",
-    "qwen-turbo",
-    "qwen2",
-    "qwen3",
-    "step-1",
-    "step-1v",
-    "step-2",
-    "tencentcloud",
-    "tongji",
-    "tongyi",
-    "tongyi-qianwen",
-    "vision-model",
-    "volcengine",
-    "wu-yuan",
-    "wuyuan",
-    "yi-ai",
-    "yi-large",
-    "yi-lightning",
-    "yi-medium",
-    "zai-glm",
-    "zai-glm-4",
-    "zhipu-ai",
-    "中国ai",
-    "中国大模型",
-    "中文chatgpt",
-    "中文embedding",
-    "中文langchain",
-    "中文llamaindex",
-    "中文nlp",
-    "中文seo",
-    "中文多模态",
-    "中文大模型",
-    "中文搜索引擎优化",
-    "中文模型",
-    "华为云",
-    "向量化",
-    "国产ai",
-    "国产大模型",
-    "多模型路由",
-    "多模态",
-    "大模型路由",
-    "字节ai",
-    "智能路由",
-    "火山引擎",
-    "百度ai",
-    "百度云",
-    "腾讯ai",
-    "腾讯云",
-    "长上下文",
-    "阿里ai",
-    "阿里云"
+    "llm-failover",
+    "mistral-proxy",
+    "openai-compatible",
+    "semantic-cache",
+    "llm"
   ],
   "author": "Das-rebel <subho@example.com>",
   "license": "MIT",
@@ -471,8 +180,7 @@
     "test": "node test.js && node test/provider-test.js",
     "test:providers": "node test/provider-test.js",
     "benchmark": "node test/benchmark.js",
-    "benchmark:verbose": "node test/benchmark.js --verbose",
-    "build": "tsc"
+    "benchmark:verbose": "node test/benchmark.js --verbose"
   },
   "engines": {
     "node": ">=18.0.0"
@@ -490,7 +198,6 @@
   },
   "devDependencies": {
     "@types/node": "^25.8.0",
-    "tsx": "^4.22.3",
     "typescript": "^6.0.3"
   }
 }