Skip to content

fix: switch default AI model to gemini-2.5-flash to avoid quota limits #288

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
omsherikar wants to merge 1 commit into
base: main
Choose a base branch
from
Open

fix: switch default AI model to gemini-2.5-flash to avoid quota limits #288

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
10 changes: 5 additions & 5 deletions apps/bubblelab-api/src/config/bubbleflow-generation-prompts.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -416,13 +416,13 @@ GOOD EXAMPLE:
\`\`\`typescript
// Searches for academic papers related to the topic variable and summarizes each one's key findings.
// The search behavior is controlled by the task prompt - modify it to focus on specific aspects,
// add date ranges, or filter by publication type. Currently using gemini-3-pro-preview for thorough
// multi-step research; switch to gemini-2.5-flash if you need faster results with less depth.
// add date ranges, or filter by publication type. Currently using gemini-2.5-flash for thorough
// multi-step research; current default is gemini-2.5-flash for balanced speed and cost.
// Returns an array of papers (each with title, url, authors, publicationDate, summary, and
// relevance explanation) plus an overallSummary that synthesizes all findings for downstream use.
const researchTool = new ResearchAgentTool({
task: \`Find research papers about \${topic}...\`,
model: 'google/gemini-3-pro-preview',
task: \`Find research papers about \${ topic }...\`,
model: 'google/gemini-2.5-flash',
expectedResultSchema: z.object({...})
});

Expand Down Expand Up @@ -500,7 +500,7 @@ MODEL SELECTION

TIER 1 - BEST (${RECOMMENDED_MODELS.BEST}):
Use for: Complex reasoning, tool-calling agents, research-agent-tool, code generation,
high-iteration tasks (50+), critical accuracy requirements
high-iteration tasks (50+), critical accuracy requirements (Note: gemini-2.5-flash is now the default for BEST)

TIER 2 - PRO (${RECOMMENDED_MODELS.PRO}, ${RECOMMENDED_MODELS.PRO_ALT}):
Use for: Multi-step reasoning, strategic planning, complex data analysis
Expand Down
4 changes: 2 additions & 2 deletions apps/bubblelab-api/src/services/ai/bubbleflow-generator.workflow.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -531,10 +531,10 @@ ${AI_AGENT_BEHAVIOR_INSTRUCTIONS}`;

model: {
reasoningEffort: 'medium',
model: 'google/gemini-3-flash-preview',
model: 'google/gemini-2.5-flash',
temperature: 0.3,
backupModel: {
model: 'anthropic/claude-sonnet-4-5',
model: 'google/gemini-2.5-flash-lite',
temperature: 0.3,
},
},
Expand Down
6 changes: 3 additions & 3 deletions packages/bubble-shared-schemas/src/bubbleflow-generation-prompts.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,11 +8,11 @@ import { AvailableModel } from './ai-models.js';

// Model constants for AI agent instructions
export const RECOMMENDED_MODELS = {
BEST: 'google/gemini-3-pro-preview',
BEST: 'google/gemini-2.5-flash',
BEST_ALT: 'openai/gpt-5.2',
PRO: 'google/gemini-3-flash-preview',
PRO: 'google/gemini-2.5-flash',
PRO_ALT: 'anthropic/claude-sonnet-4-5',
FAST: 'google/gemini-3-flash-preview',
FAST: 'google/gemini-2.5-flash',
FAST_ALT: 'anthropic/claude-haiku-4-5',
LITE: 'google/gemini-2.5-flash-lite',
IMAGE: 'google/gemini-3-pro-image-preview',
Expand Down
2 changes: 1 addition & 1 deletion packages/bubble-shared-schemas/src/coffee.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,7 +10,7 @@ import { CredentialType } from './types.js';
// Constants
export const COFFEE_MAX_ITERATIONS = 30;
export const COFFEE_MAX_QUESTIONS = 5;
export const COFFEE_DEFAULT_MODEL = 'google/gemini-3-pro-preview' as const;
export const COFFEE_DEFAULT_MODEL = 'google/gemini-2.5-flash' as const;

// ============================================================================
// Clarification Schemas
Expand Down
Loading