feat(llm): Added max context length configuration and statistics

apps/gui/next-env.d.ts

@@ -1,6 +1,6 @@
 /// <reference types="next" />
 /// <reference types="next/image-types/global" />
-import "./.next/dev/types/routes.d.ts";
+import "./.next/types/routes.d.ts";
 
 // NOTE: This file should not be edited
 // see https://nextjs.org/docs/app/api-reference/config/typescript for more information.

apps/gui/src/app/config/page.tsx

@@ -37,6 +37,7 @@ export default function ConfigPage() {
   const [editModel, setEditModel] = useState("gemini-2.5-flash");
   const [editIsActive, setEditIsActive] = useState(false);
   const [editType, setEditType] = useState<"generative" | "embedding">("generative");
+  const [editMaxContext, setEditMaxContext] = useState<number>(32768);
 
   useEffect(() => {
     if (selectedInstanceId === null) {
@@ -46,6 +47,7 @@ export default function ConfigPage() {
       setEditModel("gemini-2.5-flash");
       setEditIsActive(false);
       setEditType("generative");
+      setEditMaxContext(32768);
     } else if (selectedInstanceId === "new") {
       setEditName("");
       const defaultProvider = "google-genai";
@@ -55,6 +57,7 @@ export default function ConfigPage() {
       const pMeta = availableProviders.find((p) => p.id === defaultProvider);
       setEditModel(pMeta?.defaultModel || "gemini-2.5-flash");
       setEditIsActive(false);
+      setEditMaxContext(32768);
     } else {
       const inst = instances.find((i) => i.id === selectedInstanceId);
       if (inst) {
@@ -65,6 +68,7 @@ export default function ConfigPage() {
         const pMeta = availableProviders.find((p) => p.id === inst.providerName);
         setEditModel(inst.modelName || (inst.type === "embedding" ? pMeta?.defaultEmbeddingModel : pMeta?.defaultModel) || "gemini-2.5-flash");
         setEditIsActive(inst.isActive);
+        setEditMaxContext(inst.maxContext !== undefined && inst.maxContext !== null ? inst.maxContext : 32768);
       }
     }
   }, [selectedInstanceId, instances, availableProviders]);
@@ -132,13 +136,14 @@ export default function ConfigPage() {
           setLoading(false);
           return;
         }
-        const created = await createProviderInstance(editName, editProvider, editKey, editModel || undefined, editType);
+        const created = await createProviderInstance(editName, editProvider, editKey, editModel || undefined, editType, editType === "generative" ? editMaxContext : 0);
         if (editIsActive) {
           await setActiveProviderInstance(created.id);
         }
         targetInstanceId = created.id;
         setSelectedInstanceId(created.id);
       } else {
+        if (!selectedInstanceId) return;
         const inst = instances.find((i) => i.id === selectedInstanceId);
         if (inst && inst.type === "embedding") {
           const isMapped = mappings["embeddings"] === selectedInstanceId;
@@ -158,7 +163,7 @@ export default function ConfigPage() {
           }
         }
 
-        await updateProviderInstance(selectedInstanceId, editName, editProvider, editKey || undefined, editModel || undefined, editType);
+        await updateProviderInstance(selectedInstanceId, editName, editProvider, editKey || undefined, editModel || undefined, editType, editType === "generative" ? editMaxContext : 0);
         if (editIsActive) {
           await setActiveProviderInstance(selectedInstanceId);
         }
@@ -167,7 +172,7 @@ export default function ConfigPage() {
       await loadInstances();
       await loadMappings();
 
-      if (shouldRegenerate && targetInstanceId !== "new") {
+      if (shouldRegenerate && targetInstanceId && targetInstanceId !== "new") {
         await regenerateEmbeddings(targetInstanceId);
       }
     } catch (err) {
@@ -381,6 +386,23 @@ export default function ConfigPage() {
                         />
                       </div>
 
+                      {editType === "generative" && (
+                        <div className="flex flex-col gap-1.5">
+                          <label htmlFor="formMaxContext" className="text-xs font-medium text-gray-700">
+                            Max Context Length (Tokens, 0 for infinite)
+                          </label>
+                          <input
+                            id="formMaxContext"
+                            type="number"
+                            value={editMaxContext}
+                            onChange={(e) => setEditMaxContext(parseInt(e.target.value) || 0)}
+                            min={0}
+                            placeholder="e.g. 32768"
+                            className="w-full rounded-md border border-gray-300 bg-white px-3 py-2 text-sm outline-none transition-[border-color,box-shadow] focus:border-blue-500 focus:ring-3 focus:ring-blue-500/15"
+                          />
+                        </div>
+                      )}
+
                       <div className="mt-1 flex flex-row items-center gap-2">
                         <input
                           id="formActive"

apps/gui/src/app/play/actions.ts

@@ -243,8 +243,9 @@ export async function createProviderInstance(
   apiKey: string,
   modelName?: string,
   type: "generative" | "embedding" = "generative",
+  maxContext?: number,
 ): Promise<ModelProviderInstance> {
-  return ProviderManager.create(name, providerName, apiKey, modelName, type);
+  return ProviderManager.create(name, providerName, apiKey, modelName, type, maxContext);
 }
 
 export async function deleteProviderInstance(id: string): Promise<void> {
@@ -262,8 +263,9 @@ export async function updateProviderInstance(
   apiKey?: string,
   modelName?: string,
   type: "generative" | "embedding" = "generative",
+  maxContext?: number,
 ): Promise<void> {
-  ProviderManager.update(id, name, providerName, apiKey, modelName, type);
+  ProviderManager.update(id, name, providerName, apiKey, modelName, type, maxContext);
 }
 
 export async function getProviderMappings(): Promise<Record<string, string>> {

apps/gui/src/components/play/PlayView.tsx

@@ -13,7 +13,7 @@ import {
   listProviderInstances,
 } from "@/app/play/actions";
 import type { SimSnapshot } from "@/lib/simulation-types";
-import type { LLMProviderInstance } from "@omnia/llm";
+import type { ModelProviderInstance } from "@omnia/llm";
 
 function IntentTag({
   intent,
@@ -90,9 +90,9 @@ function PromptModal({
     const memTokens = Math.max(0, inputTokens - sysTokens - worldTokens);
 
     return [
-      { label: "System Prompt", pct: sysPct, tokens: sysTokens, type: "system", content: systemPrompt },
-      { label: "World Info", pct: worldPct, tokens: worldTokens, type: "world", content: worldStr },
-      { label: "Recent Memories", pct: memPct, tokens: memTokens, type: "memories", content: memStr || "(No memories yet.)" },
+      { label: "System Prompt", pct: sysPct, relativePct: sysPct, tokens: sysTokens, type: "system", content: systemPrompt },
+      { label: "World Info", pct: worldPct, relativePct: worldPct, tokens: worldTokens, type: "world", content: worldStr },
+      { label: "Recent Memories", pct: memPct, relativePct: memPct, tokens: memTokens, type: "memories", content: memStr || "(No memories yet.)" },
     ];
   };
 
@@ -124,15 +124,35 @@ function PromptModal({
     const proseTokens = Math.max(0, inputTokens - sysTokens - worldTokens);
 
     return [
-      { label: "System Prompt", pct: sysPct, tokens: sysTokens, type: "system", content: systemPrompt },
-      { label: "Decoder Context", pct: worldPct, tokens: worldTokens, type: "world", content: worldStr },
-      { label: "Narrative Prose", pct: prosePct, tokens: proseTokens, type: "memories", content: proseStr },
+      { label: "System Prompt", pct: sysPct, relativePct: sysPct, tokens: sysTokens, type: "system", content: systemPrompt },
+      { label: "Decoder Context", pct: worldPct, relativePct: worldPct, tokens: worldTokens, type: "world", content: worldStr },
+      { label: "Narrative Prose", pct: prosePct, relativePct: prosePct, tokens: proseTokens, type: "memories", content: proseStr },
     ];
   };
 
   const actorBreakdown = (entry.rawPrompt && entry.usage) ? parseActorPrompt(entry.rawPrompt.systemPrompt, entry.rawPrompt.userContext, entry.usage.inputTokens) : null;
   const decoderBreakdown = (entry.decoderPrompt && entry.decoderUsage) ? parseDecoderPrompt(entry.decoderPrompt.systemPrompt, entry.decoderPrompt.userContext, entry.decoderUsage.inputTokens) : null;
 
+  const actorMaxContext = entry.usage?.maxContext !== undefined ? entry.usage.maxContext : 32768;
+  const actorUsedTokens = entry.usage?.inputTokens || 0;
+  const actorUsagePctOfContext = actorMaxContext > 0 ? (actorUsedTokens / actorMaxContext) * 100 : 0;
+  const isActorAbsolute = actorMaxContext > 0 && actorUsagePctOfContext >= 20;
+
+  const scaledActorBreakdown = actorBreakdown ? actorBreakdown.map((item) => ({
+    ...item,
+    pct: isActorAbsolute ? item.relativePct * (actorUsedTokens / actorMaxContext) : item.relativePct
+  })) : null;
+
+  const decoderMaxContext = entry.decoderUsage?.maxContext !== undefined ? entry.decoderUsage.maxContext : 32768;
+  const decoderUsedTokens = entry.decoderUsage?.inputTokens || 0;
+  const decoderUsagePctOfContext = decoderMaxContext > 0 ? (decoderUsedTokens / decoderMaxContext) * 100 : 0;
+  const isDecoderAbsolute = decoderMaxContext > 0 && decoderUsagePctOfContext >= 20;
+
+  const scaledDecoderBreakdown = decoderBreakdown ? decoderBreakdown.map((item) => ({
+    ...item,
+    pct: isDecoderAbsolute ? item.relativePct * (decoderUsedTokens / decoderMaxContext) : item.relativePct
+  })) : null;
+
   useEffect(() => {
     if (!entry.rawPrompt && entry.decoderPrompt) {
       setActiveTab("decoder");
@@ -180,37 +200,57 @@ function PromptModal({
                 </div>
               )}
 
-              {actorBreakdown && (
+              {scaledActorBreakdown && (
                 <div className="prompt-breakdown-container">
                   <div style={{ display: "flex", justifyContent: "space-between", alignItems: "center", fontSize: "0.75rem", color: "#6b7280", marginBottom: "0.25rem" }}>
                     <span style={{ fontWeight: 600 }}>Input Prompt Breakdown</span>
-                    <span>Total Input Tokens: <strong>{entry.usage?.inputTokens}</strong></span>
+                    <span>
+                      Total Input Tokens: <strong>{actorUsedTokens}</strong>
+                      {actorMaxContext > 0 ? (
+                        <span> / {actorMaxContext} ({actorUsagePctOfContext.toFixed(1)}% used)</span>
+                      ) : (
+                        <span> (infinite context)</span>
+                      )}
+                    </span>
                   </div>
                   <div className="prompt-breakdown-bar">
-                    {actorBreakdown.map((item, idx) => (
+                    {scaledActorBreakdown.map((item, idx) => {
+                      const displayPct = actorMaxContext > 0 ? (item.tokens / actorMaxContext) * 100 : item.relativePct;
+                      return (
+                        <div
+                          key={idx}
+                          className={`bar-section ${item.type}`}
+                          style={{ width: `${item.pct}%` }}
+                          title={`${item.label}: ${item.tokens} tokens (${displayPct.toFixed(1)}%)`}
+                        />
+                      );
+                    })}
+                    {isActorAbsolute && (
                       <div
-                        key={idx}
-                        className={`bar-section ${item.type}`}
-                        style={{ width: `${item.pct}%` }}
-                        title={`${item.label}: ${item.tokens} tokens (${item.pct.toFixed(1)}%)`}
+                        className="bar-section-empty"
+                        style={{ width: `${100 - actorUsagePctOfContext}%`, backgroundColor: "#ffffff", height: "100%" }}
+                        title={`Available: ${actorMaxContext - actorUsedTokens} tokens (${(100 - actorUsagePctOfContext).toFixed(1)}% remaining)`}
                       />
-                    ))}
+                    )}
                   </div>
                   <div className="breakdown-accordion">
-                    {actorBreakdown.map((item, idx) => (
-                      <details key={idx} className="breakdown-accordion-item" open={idx === 0}>
-                        <summary className="accordion-header">
-                          <span className={`legend-color ${item.type}`} />
-                          <span className="header-text">
-                            {item.label}: <strong>{item.tokens}</strong> tokens ({item.pct.toFixed(0)}%)
-                          </span>
-                          <span className="accordion-chevron">▼</span>
-                        </summary>
-                        <div className="accordion-content">
-                          <pre>{item.content}</pre>
-                        </div>
-                      </details>
-                    ))}
+                    {scaledActorBreakdown.map((item, idx) => {
+                      const displayPct = actorMaxContext > 0 ? (item.tokens / actorMaxContext) * 100 : item.relativePct;
+                      return (
+                        <details key={idx} className="breakdown-accordion-item" open={idx === 0}>
+                          <summary className="accordion-header">
+                            <span className={`legend-color ${item.type}`} />
+                            <span className="header-text">
+                              {item.label}: <strong>{item.tokens}</strong> tokens ({displayPct.toFixed(0)}%)
+                            </span>
+                            <span className="accordion-chevron">▼</span>
+                          </summary>
+                          <div className="accordion-content">
+                            <pre>{item.content}</pre>
+                          </div>
+                        </details>
+                      );
+                    })}
                   </div>
                 </div>
               )}
@@ -240,37 +280,57 @@ function PromptModal({
                 </div>
               )}
 
-              {decoderBreakdown && (
+              {scaledDecoderBreakdown && (
                 <div className="prompt-breakdown-container">
                   <div style={{ display: "flex", justifyContent: "space-between", alignItems: "center", fontSize: "0.75rem", color: "#6b7280", marginBottom: "0.25rem" }}>
                     <span style={{ fontWeight: 600 }}>Input Prompt Breakdown</span>
-                    <span>Total Input Tokens: <strong>{entry.decoderUsage?.inputTokens}</strong></span>
+                    <span>
+                      Total Input Tokens: <strong>{decoderUsedTokens}</strong>
+                      {decoderMaxContext > 0 ? (
+                        <span> / {decoderMaxContext} ({decoderUsagePctOfContext.toFixed(1)}% used)</span>
+                      ) : (
+                        <span> (infinite context)</span>
+                      )}
+                    </span>
                   </div>
                   <div className="prompt-breakdown-bar">
-                    {decoderBreakdown.map((item, idx) => (
+                    {scaledDecoderBreakdown.map((item, idx) => {
+                      const displayPct = decoderMaxContext > 0 ? (item.tokens / decoderMaxContext) * 100 : item.relativePct;
+                      return (
+                        <div
+                          key={idx}
+                          className={`bar-section ${item.type}`}
+                          style={{ width: `${item.pct}%` }}
+                          title={`${item.label}: ${item.tokens} tokens (${displayPct.toFixed(1)}%)`}
+                        />
+                      );
+                    })}
+                    {isDecoderAbsolute && (
                       <div
-                        key={idx}
-                        className={`bar-section ${item.type}`}
-                        style={{ width: `${item.pct}%` }}
-                        title={`${item.label}: ${item.tokens} tokens (${item.pct.toFixed(1)}%)`}
+                        className="bar-section-empty"
+                        style={{ width: `${100 - decoderUsagePctOfContext}%`, backgroundColor: "#ffffff", height: "100%" }}
+                        title={`Available: ${decoderMaxContext - decoderUsedTokens} tokens (${(100 - decoderUsagePctOfContext).toFixed(1)}% remaining)`}
                       />
-                    ))}
+                    )}
                   </div>
                   <div className="breakdown-accordion">
-                    {decoderBreakdown.map((item, idx) => (
-                      <details key={idx} className="breakdown-accordion-item" open={idx === 0}>
-                        <summary className="accordion-header">
-                          <span className={`legend-color ${item.type}`} />
-                          <span className="header-text">
-                            {item.label}: <strong>{item.tokens}</strong> tokens ({item.pct.toFixed(0)}%)
-                          </span>
-                          <span className="accordion-chevron">▼</span>
-                        </summary>
-                        <div className="accordion-content">
-                          <pre>{item.content}</pre>
-                        </div>
-                      </details>
-                    ))}
+                    {scaledDecoderBreakdown.map((item, idx) => {
+                      const displayPct = decoderMaxContext > 0 ? (item.tokens / decoderMaxContext) * 100 : item.relativePct;
+                      return (
+                        <details key={idx} className="breakdown-accordion-item" open={idx === 0}>
+                          <summary className="accordion-header">
+                            <span className={`legend-color ${item.type}`} />
+                            <span className="header-text">
+                              {item.label}: <strong>{item.tokens}</strong> tokens ({displayPct.toFixed(0)}%)
+                            </span>
+                            <span className="accordion-chevron">▼</span>
+                          </summary>
+                          <div className="accordion-content">
+                            <pre>{item.content}</pre>
+                          </div>
+                        </details>
+                      );
+                    })}
                   </div>
                 </div>
               )}
@@ -488,7 +548,7 @@ export function PlayView() {
   const [availableEntities, setAvailableEntities] = useState<{ id: string; name: string }[]>([]);
   const [selectedEntity, setSelectedEntity] = useState("");
 
-  const [providerInstances, setProviderInstances] = useState<LLMProviderInstance[]>([]);
+  const [providerInstances, setProviderInstances] = useState<ModelProviderInstance[]>([]);
 
   // Load scenarios and provider instances on mount
   useEffect(() => {

apps/gui/src/lib/simulation-types.ts

@@ -26,6 +26,7 @@ export interface LogEntry {
     totalTokens: number;
     modelName?: string;
     providerInstanceName?: string;
+    maxContext?: number;
   };
   decoderPrompt?: {
     systemPrompt: string;
@@ -37,6 +38,7 @@ export interface LogEntry {
     totalTokens: number;
     modelName?: string;
     providerInstanceName?: string;
+    maxContext?: number;
   };
 }
 

apps/gui/src/lib/simulation.ts

@@ -235,11 +235,12 @@ class SimulationManager {
       const providerName = inst ? inst.providerName : "google-genai";
       const modelName = inst ? inst.modelName : undefined;
       const instanceName = inst ? inst.name : undefined;
+      const maxContext = inst ? inst.maxContext : undefined;
 
       if (providerName === "google-genai") {
-        return new GeminiProvider(key, modelName, instanceName);
+        return new GeminiProvider(key, modelName, instanceName, maxContext);
       } else if (providerName === "openrouter") {
-        return new OpenRouterProvider(key, modelName, instanceName);
+        return new OpenRouterProvider(key, modelName, instanceName, maxContext);
       } else {
         return new MockLLMProvider([]);
       }
@@ -702,9 +703,9 @@ class SimulationManager {
         }
 
         if (inst.providerName === "google-genai") {
-          return new GeminiProvider(inst.apiKey, inst.modelName, inst.name);
+          return new GeminiProvider(inst.apiKey, inst.modelName, inst.name, inst.maxContext);
         } else if (inst.providerName === "openrouter") {
-          return new OpenRouterProvider(inst.apiKey, inst.modelName, inst.name);
+          return new OpenRouterProvider(inst.apiKey, inst.modelName, inst.name, inst.maxContext);
         } else {
           return new MockLLMProvider([]);
         }