Skip to content

Feat/final synthesis compaction #1273

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
Aayushjshah wants to merge 3 commits into
base: main
Choose a base branch
from
Open

Feat/final synthesis compaction #1273

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
78d42c6
feat: compaction in final Answer generation
aayushJuspay Mar 8, 2026
9199f39
feat: ls comment changes for aborting flow on stop command from user
aayushJuspay Mar 9, 2026
03e9341
feat: fetching maxInputTokens from model info api itself
aayushJuspay Mar 9, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
969 changes: 622 additions & 347 deletions server/ai/fetchModels.ts
View file
Open in desktop

Large diffs are not rendered by default.

11 changes: 8 additions & 3 deletions server/ai/provider/base.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,7 @@
import { type Message } from "@aws-sdk/client-bedrock-runtime"
import type { ConverseResponse, LLMProvider, ModelParams } from "@/ai/types"
import { AIProviders } from "@/ai/types"
import { getEffectiveMaxOutputTokens } from "@/ai/fetchModels"
import { MODEL_CONFIGURATIONS } from "@/ai/modelConfig"
import config from "@/config"
import path from "path"
Expand All @@ -19,10 +20,14 @@ abstract class Provider implements LLMProvider {
getModelParams(params: ModelParams) {
// Look up the actual model name from MODEL_CONFIGURATIONS
// This resolves enum values like "vertex-claude-sonnet-4" to actual API model names like "claude-sonnet-4@20250514"
const modelConfig = MODEL_CONFIGURATIONS[params.modelId || defaultFastModel]
const actualModelId = modelConfig?.actualName || params.modelId || defaultFastModel
const resolvedModelId = params.modelId || defaultFastModel
const modelConfig = MODEL_CONFIGURATIONS[resolvedModelId]
const actualModelId = modelConfig?.actualName || resolvedModelId || defaultFastModel
const requestedMaxTokens = params.max_new_tokens || 1024 * 8
return {
maxTokens: params.max_new_tokens || 1024 * 8,
maxTokens:
getEffectiveMaxOutputTokens(resolvedModelId, requestedMaxTokens) ??
requestedMaxTokens,
topP: params.top_p || 0.9,
temperature: params.temperature || 0.6,
modelId: actualModelId || defaultFastModel,
Expand Down
Loading