Skip to content

Commit 7a06640

Browse files
author
FastJava Team
committed
docs: consolidate AI module architecture and roadmap
1 parent 112b3c0 commit 7a06640

2 files changed

Lines changed: 31 additions & 126 deletions

File tree

‎README.md‎

Lines changed: 15 additions & 64 deletions
Original file line numberDiff line numberDiff line change
@@ -125,73 +125,24 @@ A unified directory of all FastJava modules. Categorized by functional domain, i
125125
| FastKeychain | — | Credential store | Windows | TODO | — |
126126

127127
### 12. AI
128-
129-
#### AI Brain
130128
| Module | Speed vs Java | Key Metrics | Tech | Status | Version |
131129
|--------|---------------|-------------|------|--------|---------|
132-
| **[FastAI](https://github.com/andrestubbe/FastAI)** | **—** | Provider-Router (lokal/Cloud) | API | Alpha | `v0.1.5` |
133-
| **[FastAIAgent](https://github.com/andrestubbe/FastAIAgent)** | **—** | Agenten-Logik, Tool-Auswahl | LLM | Alpha | `v0.1.5` |
130+
| **[FastAI](https://github.com/andrestubbe/FastAI)** | **—** | Unified provider routing & LLM client layer | Multi-Provider | Alpha | `v0.1.5` |
131+
| **[FastAIAgent](https://github.com/andrestubbe/FastAIAgent)** | **—** | Autonomous agent logic, tools & planning | Agent Loop | Alpha | `v0.1.5` |
134132
| **[FastAIBot](https://github.com/andrestubbe/FastAIBot)** | **—** | High-performance bot orchestrator | LLM / Runtime | Alpha | `v0.1.2` |
135-
| **[FastAIRuntime](https://github.com/andrestubbe/FastAIRuntime)** | **—** | Deterministic execution engine | Engine | Alpha | `v0.1.0` |
136-
| **FastAIReasoner** | — | Planning & Multi-Step-Reasoning | Logic | TODO | — |
137-
| **FastAIWorkflow** | — | Multi-Agent-Pipelines & DAG | Orchestrator | TODO | — |
138-
| **FastAIState** | — | Shared Agent State / Blackboard | Memory | TODO | — |
139-
140-
#### AI Memory
141-
| Module | Speed vs Java | Key Metrics | Tech | Status | Version |
142-
|--------|---------------|-------------|------|--------|---------|
143-
| **FastAISession** | — | Conversation Lifecycle | History | TODO | — |
144-
| **FastAIContext** | — | Kontext-Management, Rollen | State | TODO | — |
145-
| **[FastAIMemory](https://github.com/andrestubbe/FastAIMemory)** | **—** | Kurzzeit- & Langzeit-Gedächtnis | Context / History | Alpha | `v0.1.3` |
146-
| **[FastAIVectorDB](https://github.com/andrestubbe/FastAIVectorDB)** | **—** | Vektor-Store für Memory & RAG | SIMD | Alpha | `v0.1.4` |
147-
| **FastAIEmbedding** | — | Embedding-Service | Model | TODO | — |
148-
| **[FastAIRag](https://github.com/andrestubbe/FastAIRag)** | **—** | Retrieval-Pipeline | RAG | Alpha | `v0.1.1` |
149-
150-
#### AI Engine
151-
| Module | Speed vs Java | Key Metrics | Tech | Status | Version |
152-
|--------|---------------|-------------|------|--------|---------|
153-
| **[FastAIModel](https://github.com/andrestubbe/FastAIModel)** | **—** | Lokale Inferenz-Runtime (gguf/ONNX) | ONNX | Alpha | `v0.1.2` |
154-
| **[FastContentChunk](https://github.com/andrestubbe/FastContentChunk)** | **—** | Tokenizer & chunking strategy engine | Tokenizer | Alpha | `v0.1.3` |
155-
| **FastAIStream** | — | Token-Streaming-Engine | Pipeline | TODO | — |
156-
| **FastAIToken** | — | Tokenization, Counting, Budgeting | Tokenizer | TODO | — |
157-
| **FastAICache** | — | Antwort- & Embedding-Caching | Cache | TODO | — |
158-
159-
#### AI Hands
160-
| Module | Speed vs Java | Key Metrics | Tech | Status | Version |
161-
|--------|---------------|-------------|------|--------|---------|
162-
| **FastAIToolBridge**| — | Tool-Ausführungsschicht | API | TODO | — |
163-
| **FastAIMCP** | — | Model Context Protocol Adapter | MCP | TODO | — |
164-
| **FastAISandbox** | — | Code-Isolation | Security | TODO | — |
165-
| **FastAIScheduler** | — | Async Jobs, Retries, Cron | Queueing | TODO | — |
166-
167-
#### AI Guards
168-
| Module | Speed vs Java | Key Metrics | Tech | Status | Version |
169-
|--------|---------------|-------------|------|--------|---------|
170-
| **FastAIGuard** | — | Prompt-Injection-Schutz | Security | TODO | — |
171-
| **FastAIAuth** | — | API-Keys, User-Permissions | Auth | TODO | — |
172-
| **FastAIPolicy** | — | Declarative AI Rules / Policies | Rules | TODO | — |
173-
174-
#### AI Dashboard
175-
| Module | Speed vs Java | Key Metrics | Tech | Status | Version |
176-
|--------|---------------|-------------|------|--------|---------|
177-
| **FastAITrace** | — | Token-Flow & Tool-Tracing | Debugging | TODO | — |
178-
| **FastAIEval** | — | Automatische Antwort-Evaluation | Eval | TODO | — |
179-
| **FastAIMetrics** | — | Performance- & Usage-Metrics | Telemetry | TODO | — |
180-
| **FastAIReplay** | — | Replay von Sessions | Debugging | TODO | — |
181-
| **FastAIParser** | — | Strukturierter Output-Parser | Parser | TODO | — |
182-
| **FastAIPrompt** | — | Prompt-Compiler, Templates | Compiler | TODO | — |
183-
| **FastAIRegistry** | — | Registry für Models, Tools, Agents| Registry | TODO | — |
184-
| **FastAIPlugin** | — | Plugin-/Extension-System | Plugin | TODO | — |
185-
| **FastEventBus** | — | Event-System für Tokens & Tools | Event Bus | TODO | — |
186-
187-
#### AI Infrastructure
188-
| Module | Speed vs Java | Key Metrics | Tech | Status | Version |
189-
|--------|---------------|-------------|------|--------|---------|
190-
| **FastAPI** | — | REST/WebSocket/gRPC Layer | Network | TODO | — |
191-
| **FastConfig** | — | Runtime-/Model-Konfiguration | Config | TODO | — |
192-
| **FastArtifact** | — | Verwaltung generierter Dateien | Storage | TODO | — |
193-
| **FastStorage** | — | Files, Blobs, Transcripts | IO | TODO | — |
194-
| **FastAICluster** | — | Multi-GPU / Multi-Node | Orchestrator | TODO | — |
133+
| **[FastAIRuntime](https://github.com/andrestubbe/FastAIRuntime)** | **—** | Deterministic execution engine & FastAIEventBus | Runtime | Alpha | `v0.1.0` |
134+
| **[FastAIModel](https://github.com/andrestubbe/FastAIModel)** | **—** | Local ONNX & GGUF matrix inference runtime | ONNX / DirectML | Alpha | `v0.1.2` |
135+
| **[FastAIRag](https://github.com/andrestubbe/FastAIRag)** | **—** | High-speed retrieval augmented generation pipeline | RAG / Hybrid | Alpha | `v0.1.1` |
136+
| **[FastAIMemory](https://github.com/andrestubbe/FastAIMemory)** | **—** | Short-term & long-term conversational memory | Context Store | Alpha | `v0.1.3` |
137+
| **[FastAIVectorDB](https://github.com/andrestubbe/FastAIVectorDB)** | **—** | Zero-allocation vector database with built-in Embedder | AVX2 / JNI | Alpha | `v0.1.4` |
138+
| **[FastAIMCP](https://github.com/andrestubbe/FastAIMCP)** | **—** | Model Context Protocol client & stdio tool bridge | MCP | Alpha | `v0.1.0` |
139+
| **[FastAIReasoner](https://github.com/andrestubbe/FastAIReasoner)** | **—** | Multi-step reasoning, CoT & heuristic planning | Tree-of-Thoughts | Alpha | `v0.1.0` |
140+
| **[FastAIGraph](https://github.com/andrestubbe/FastAIGraph)** | **—** | In-memory knowledge graph & multi-hop traversal | Entity Linking | Alpha | `v0.1.0` |
141+
| **[FastAIRerank](https://github.com/andrestubbe/FastAIRerank)** | **—** | Cross-encoder relevance scoring & ranking | Reranker | Alpha | `v0.1.0` |
142+
| **[FastAIHybrid](https://github.com/andrestubbe/FastAIHybrid)** | **—** | Hybrid dense & sparse retrieval fusion | Fusion Engine | Alpha | `v0.1.0` |
143+
| FastAIState | — | Lock-free shared agent state & blackboard memory | Shared Memory | Planned | — |
144+
| FastAIMetrics | — | Token usage, latency profiling, cost & evaluation | Telemetry | Planned | — |
145+
| FastVision | — | Local GPU multimodal vision & UI grounding pipeline | DirectML / Vulkan | Planned | — |
195146

196147
### 13. Vision
197148
| Module | Speed vs Java | Key Metrics | Tech | Status | Version |

‎docs/MAP.md‎

Lines changed: 16 additions & 62 deletions
Original file line numberDiff line numberDiff line change
@@ -106,70 +106,24 @@
106106
| FastSTT | Speech-to-text | Whisper/Cloud |
107107

108108
### AI & ML
109-
110-
#### AI Brain
111-
| Module | Purpose | Tech |
112-
|--------|---------|------|
113-
| FastAI | Provider-Router (lokal/Cloud) | API |
114-
| FastAIAgent | Agenten-Logik, Tool-Auswahl | LLM |
115-
| FastAIReasoner | Planning, Reflection | Logic |
116-
| FastAIWorkflow | Multi-Agent-Pipelines, Tasks | Orchestrator |
117-
| FastAIState | Shared Agent State / Blackboard | Memory |
118-
119-
#### AI Memory
120-
| Module | Purpose | Tech |
121-
|--------|---------|------|
122-
| FastAISession | Conversation Lifecycle | History |
123-
| FastAIContext | Kontext-Management, Rollen | State |
124-
| FastAIMemory | Kurzzeit- & Langzeit-Gedächtnis | Vector DB |
125-
| FastAIVectorDB| Vektor-Store für Memory & RAG | SIMD |
126-
| FastAIEmbedding| Embedding-Service | Model |
127-
| FastAIRAG | Retrieval-Pipeline | RAG |
128-
129-
#### AI Engine
130-
| Module | Purpose | Tech |
131-
|--------|---------|------|
132-
| FastModel | Lokale Inferenz-Runtime (gguf/ONNX) | ONNX |
133-
| FastAIStream | Token-Streaming-Engine | Pipeline |
134-
| FastAIToken | Tokenization, Counting, Budgeting | Tokenizer |
135-
| FastAICache | Antwort- & Embedding-Caching | Cache |
136-
137-
#### AI Hands
138-
| Module | Purpose | Tech |
139-
|--------|---------|------|
140-
| FastAIToolBridge| Tool-Ausführungsschicht | API |
141-
| FastAIMCP | Model Context Protocol Adapter | MCP |
142-
| FastAISandbox | Code-Isolation | Security |
143-
| FastAIScheduler| Async Jobs, Retries, Cron | Queueing |
144-
145-
#### AI Guards
146-
| Module | Purpose | Tech |
147-
|--------|---------|------|
148-
| FastAIGuard | Prompt-Injection-Schutz | Security |
149-
| FastAIAuth | API-Keys, User-Permissions | Auth |
150-
| FastAIPolicy | Declarative AI Rules / Policies | Rules |
151-
152-
#### AI Dashboard
153-
| Module | Purpose | Tech |
154-
|--------|---------|------|
155-
| FastAITrace | Token-Flow, Tool-Tracing | Debugging |
156-
| FastAIEval | Automatische Antwort-Evaluation | Eval |
157-
| FastAIMetrics | Performance-, Latency-Metrics | Telemetry |
158-
| FastAIReplay | Replay von Sessions | Debugging |
159-
| FastAIParser | Strukturierter Output-Parser | Parser |
160-
| FastAIPrompt | Prompt-Compiler, Templates | Compiler |
161-
| FastAIRegistry| Registry für Models, Tools, Agents| Registry |
162-
| FastAIPlugin | Plugin-/Extension-System | Plugin |
163-
| FastEventBus | Event-System für Tokens | Event Bus |
164-
165-
#### AI Infrastructure
166109
| Module | Purpose | Tech |
167110
|--------|---------|------|
168-
| FastAPI | REST/WebSocket/gRPC Layer | Network |
169-
| FastConfig | Zentrale Runtime-Konfiguration | Config |
170-
| FastArtifact | Verwaltung von Dateien, Images | Storage |
171-
| FastStorage | Files, Blobs, Transcripts | IO |
172-
| FastAICluster | Multi-GPU / Multi-Node | Orchestrator |
111+
| FastAI | Multi-provider router & LLM client layer | Multi-Provider API |
112+
| FastAIAgent | Autonomous agent logic, tools & planning loop | Agent Runtime |
113+
| FastAIBot | High-performance bot orchestrator & session manager | LLM Engine |
114+
| FastAIRuntime | Deterministic execution engine & FastAIEventBus | Runtime / Events |
115+
| FastAIModel | Local ONNX/GGUF matrix inference runtime | ONNX / DirectML |
116+
| FastAIRag | High-speed retrieval augmented generation pipeline | RAG / Hybrid |
117+
| FastAIMemory | Short-term & long-term conversational memory | Context Store |
118+
| FastAIVectorDB | Zero-allocation vector store with built-in Embedder | AVX2 / JNI |
119+
| FastAIMCP | Model Context Protocol client & stdio tool bridge | MCP |
120+
| FastAIReasoner | Multi-step reasoning, CoT & heuristic planning | Tree-of-Thoughts |
121+
| FastAIGraph | In-memory knowledge graph & multi-hop traversal | Entity Linking |
122+
| FastAIRerank | Cross-encoder relevance scoring & ranking | Reranker |
123+
| FastAIHybrid | Hybrid dense & sparse retrieval fusion | Fusion Engine |
124+
| FastAIState | Lock-free shared agent state & blackboard memory | Shared Memory |
125+
| FastAIMetrics | Token usage, latency profiling, cost & evaluation | Telemetry |
126+
| FastVision | Local GPU multimodal vision & UI grounding pipeline | DirectML / Vulkan |
173127

174128
### Debugging & Development
175129
| Module | Purpose | Tech |

0 commit comments

Comments
 (0)