feat: quadrant-based model routing with per-slot provider cascades
Some checks failed
Deploy (Gitea) / deploy (push) Failing after 2s
Some checks failed
Deploy (Gitea) / deploy (push) Failing after 2s
This commit is contained in:
@@ -129,4 +129,3 @@ FN receives (signal) and returns T if consumed, nil to continue."
|
||||
(test test-depth-limiting
|
||||
(let ((runaway-signal (list :type :EVENT :depth 11 :payload (list :sensor :heartbeat))))
|
||||
(is (null (process-signal runaway-signal)))))
|
||||
(defun bad-code () (broken
|
||||
|
||||
@@ -20,20 +20,23 @@
|
||||
(let ((backend-fn (gethash backend *backend-registry*)))
|
||||
(when backend-fn
|
||||
(log-message "PROBABILISTIC: Attempting backend ~a..." backend)
|
||||
(let* ((model (when *model-selector*
|
||||
(funcall *model-selector* backend context)))
|
||||
(result (if model
|
||||
(funcall backend-fn prompt system-prompt :model model)
|
||||
(funcall backend-fn prompt system-prompt))))
|
||||
(cond ((and (listp result) (eq (getf result :status) :success))
|
||||
(return (getf result :content)))
|
||||
((stringp result)
|
||||
(return result))
|
||||
(t
|
||||
(log-message "PROBABILISTIC: Backend ~a failed: ~a"
|
||||
backend (getf result :message))))))))
|
||||
(let* ((model-val (when *model-selector*
|
||||
(funcall *model-selector* backend context))))
|
||||
(if (eq model-val :skip)
|
||||
(log-message "PROBABILISTIC: Skipping ~a (filtered)" backend)
|
||||
(let* ((model (if model-val model-val nil))
|
||||
(result (if model
|
||||
(funcall backend-fn prompt system-prompt :model model)
|
||||
(funcall backend-fn prompt system-prompt))))
|
||||
(cond ((and (listp result) (eq (getf result :status) :success))
|
||||
(return (getf result :content)))
|
||||
((stringp result)
|
||||
(return result))
|
||||
(t
|
||||
(log-message "PROBABILISTIC: Backend ~a failed: ~a"
|
||||
backend (getf result :message)))))))))))
|
||||
(list :type :LOG
|
||||
:payload (list :text "Neural Cascade Failure: All providers exhausted.")))))
|
||||
:payload (list :text "Neural Cascade Failure: All providers exhausted."))))
|
||||
|
||||
(defun markdown-strip (text)
|
||||
(if (and text (stringp text))
|
||||
|
||||
88
lisp/system-model-router.lisp
Normal file
88
lisp/system-model-router.lisp
Normal file
@@ -0,0 +1,88 @@
|
||||
(defvar *model-cascade-code* nil
|
||||
"Cascade for :code tasks: ((:ollama . \"model\") ...)")
|
||||
|
||||
(defvar *model-cascade-plan* nil
|
||||
"Cascade for :plan tasks.")
|
||||
|
||||
(defvar *model-cascade-chat* nil
|
||||
"Cascade for :chat tasks.")
|
||||
|
||||
(defvar *model-cascade-background* nil
|
||||
"Cascade for background tasks (heartbeat, delegation).")
|
||||
|
||||
(defvar *local-backends* '(:ollama :llama-cpp)
|
||||
"Backend keywords considered local (privacy-safe).")
|
||||
|
||||
(defun model-classify-complexity (text)
|
||||
"Classify TEXT into :code, :plan, or :chat."
|
||||
(let ((lower (string-downcase text)))
|
||||
(cond
|
||||
((or (search "defun" lower) (search "defmacro" lower)
|
||||
(search "write" lower) (search "refactor" lower)
|
||||
(search "fix " lower) (search "implement" lower)
|
||||
(search "code" lower)
|
||||
(search "#+begin_src" lower))
|
||||
:code)
|
||||
((or (search "plan" lower) (search "roadmap" lower)
|
||||
(search "strategy" lower) (search "design" lower)
|
||||
(search "architecture" lower))
|
||||
:plan)
|
||||
(t :chat))))
|
||||
|
||||
(defun model-cascade-find (cascade backend)
|
||||
"Find first (PROVIDER . MODEL) in CASCADE matching BACKEND."
|
||||
(assoc backend cascade
|
||||
:test (lambda (a b) (string-equal (string a) (string b)))))
|
||||
|
||||
(defun model-select (backend context)
|
||||
"Select model for BACKEND given CONTEXT signal.
|
||||
Returns model name or :skip."
|
||||
(let* ((payload (getf context :payload))
|
||||
(text (or (getf payload :text) ""))
|
||||
(sensor (getf payload :sensor))
|
||||
(has-personal (and (boundp '*dispatcher-privacy-tags*)
|
||||
(some (lambda (tag) (search tag text))
|
||||
(symbol-value '*dispatcher-privacy-tags*))))
|
||||
(is-local (member backend *local-backends*)))
|
||||
;; Privacy: skip cloud backends for personal content
|
||||
(when (and has-personal (not is-local))
|
||||
(log-message "MODEL-ROUTER: Skipping ~a (personal content)" backend)
|
||||
(return-from model-select :skip))
|
||||
;; Quadrant: background tasks use background cascade
|
||||
(if (member sensor '(:heartbeat :delegation :tool-output :loop-error))
|
||||
(let ((entry (car (or *model-cascade-background*
|
||||
'((:ollama . "phi-2"))))))
|
||||
(cdr entry))
|
||||
;; Foreground: classify complexity, use slot cascade
|
||||
(let* ((slot (model-classify-complexity text))
|
||||
(cascade (case slot
|
||||
(:code *model-cascade-code*)
|
||||
(:plan *model-cascade-plan*)
|
||||
(t *model-cascade-chat*)))
|
||||
(entry (model-cascade-find
|
||||
(or cascade '((:ollama . "qwen2.5:14b"))) backend)))
|
||||
(if entry (cdr entry) :skip)))))
|
||||
|
||||
(defun model-router-init ()
|
||||
"Read env vars and wire model-select into *model-selector*."
|
||||
(flet ((parse-cascade (str)
|
||||
(when (and str (> (length str) 0))
|
||||
(let ((*read-eval* nil))
|
||||
(read-from-string str)))))
|
||||
(setf *model-cascade-code* (parse-cascade (uiop:getenv "MODEL_CASCADE_CODE"))
|
||||
*model-cascade-plan* (parse-cascade (uiop:getenv "MODEL_CASCADE_PLAN"))
|
||||
*model-cascade-chat* (parse-cascade (uiop:getenv "MODEL_CASCADE_CHAT"))
|
||||
*model-cascade-background* (parse-cascade (uiop:getenv "MODEL_CASCADE_BACKGROUND"))
|
||||
*local-backends* (let ((env (uiop:getenv "LOCAL_BACKENDS")))
|
||||
(if env
|
||||
(mapcar (lambda (s) (intern (string-upcase (string-trim " " s)) :keyword))
|
||||
(uiop:split-string env :separator '(#\,)))
|
||||
'(:ollama :llama-cpp)))))
|
||||
(setf *model-selector* #'model-select)
|
||||
(log-message "MODEL-ROUTER: Initialized, selector=~a" *model-selector*))
|
||||
|
||||
(defskill :passepartout-model-router
|
||||
:priority 250
|
||||
:trigger (lambda (ctx) (declare (ignore ctx)) nil))
|
||||
|
||||
(model-router-init)
|
||||
Reference in New Issue
Block a user