How mode-io/vllm-jev works
vLLM Jev serves structured Choice, Noul, and Score decisions from Jev-style checkpoints. The main workflow is a user starting a supported model, then submitting a decision request to an HTTP endpoint, which dispatches to the checkpoint-specific inference service and returns candidate scores and a selected label. Linux uses native vLLM; Apple Silicon has separate serving paths. The graph also preserves documented multimodal and model-family capabilities. Some subsystem wiring is inferred only from the SOURCE INDEX references, so uncertain links are omitted.
Main entry points: endpoint.py HTTP endpoint · checkpoint.py Checkpoint verification · valen.py Valen inference · media.py Media loading
Components 23
Serving
- Serve CLI
vllm_jev/cli.py - HTTP endpoint
vllm_jev/endpoint.py - Mac serving
vllm_jev/mac.py - Mac Laya
vllm_jev/mac_laya.py - Mac Valen
vllm_jev/mac_valen.py
Decision models
- Decision protocols
vllm_jev/decision_protocols.py - Typed decisions
vllm_jev/decision.py - Laya inference
vllm_jev/laya.py - Laya readout
vllm_jev/laya_graph.py
Multimodal models
- Media loading
vllm_jev/media.py - Video loading
vllm_jev/video.py - Valen inference
vllm_jev/valen.py - Clef inference
vllm_jev/clef.py - RSI-Jev inference
vllm_jev/rsijev.py - vjev inference
vllm_jev/vjev.py
Checkpoint preparation
- Checkpoint verification
vllm_jev/checkpoint.py - Decision export
vllm_jev/decision_export.py - Laya export
vllm_jev/laya_export.py - Clef export
vllm_jev/clef_export.py - RSI-Jev export
vllm_jev/rsijev_export.py - Valen export
vllm_jev/valen_export.py - vjev export
vllm_jev/vjev_export.py
Other components
- Model user
Connections 27
- Model user → Serve CLI: starts model
README.md - Model user → HTTP endpoint: submits decisions
README.md - HTTP endpoint → Typed decisions: dispatches
vllm_jev/endpoint.py - HTTP endpoint → Laya inference: dispatches
vllm_jev/endpoint.py - HTTP endpoint → Valen inference: dispatches
vllm_jev/endpoint.py - HTTP endpoint → Clef inference: dispatches
vllm_jev/endpoint.py - HTTP endpoint → RSI-Jev inference: dispatches
vllm_jev/endpoint.py - HTTP endpoint → vjev inference: dispatches
vllm_jev/endpoint.py - HTTP endpoint → Decision protocols: uses
vllm_jev/decision.py - Laya inference → Laya readout: uses readout
vllm_jev/laya.py - Mac serving → HTTP endpoint: attaches router
vllm_jev/mac.py - Mac Laya → Mac serving: uses serving
vllm_jev/mac_laya.py - Mac Valen → Mac serving: uses serving
vllm_jev/mac_valen.py - Mac Laya → Laya inference: uses inference
vllm_jev/mac_laya.py - Mac Valen → Valen inference: uses inference
vllm_jev/mac_valen.py - Valen inference → Media loading: loads images
vllm_jev/valen.py - Valen inference → Video loading: loads video
vllm_jev/valen.py - Clef inference → Media loading: loads media
vllm_jev/clef.py - Clef inference → Video loading: loads video
vllm_jev/clef.py - RSI-Jev inference → Media loading: loads images
vllm_jev/rsijev.py - vjev inference → Media loading: loads images
vllm_jev/vjev.py - Decision export → Checkpoint verification: verifies checkpoint
vllm_jev/decision_export.py - Laya export → Checkpoint verification: uses verification
vllm_jev/laya_export.py - Clef export → Checkpoint verification: uses verification
vllm_jev/clef_export.py - RSI-Jev export → Checkpoint verification: uses verification
vllm_jev/rsijev_export.py - Valen export → Checkpoint verification: uses verification
vllm_jev/valen_export.py - vjev export → Checkpoint verification: uses verification
vllm_jev/vjev_export.py