Skip to main content

​
Gemini Integration

How Simili Bot uses Google Gemini for AI analysis.

​
Services

Simili Bot uses Gemini for:
  1. Text Embeddings - Convert text to vectors
  2. LLM Analysis - AI reasoning and classification

​
Embeddings

​
text-embedding-004

Default embedding model: 
Input: Text string (up to ~4000 tokens)
Output: 768-dimensional vector
Speed: 100-500ms per request
Batch Size: Up to 100 texts
Use Cases:
  • Convert issue text to vector for similarity search
  • Generate embeddings for all issues during indexing
Cost: ~$0.025 per million input tokens

​
LLM Analysis

Uses Gemini for analysis tasks:

​
1. Duplicate Detection

Input: Current issue + similar issues
Task: Determine if duplicates
Output: Boolean + confidence (0.0-1.0)
Speed: 2-5 seconds

​
2. Quality Assessment

Input: Issue title + body
Task: Evaluate description quality
Output: Score (0-100) + suggestions
Speed: 1-3 seconds

​
3. Issue Routing

Input: Issue + repository descriptions
Task: Determine correct repository
Output: Target repo name
Speed: 2-5 seconds

​
4. Auto Triage

Input: Issue content + available labels
Task: Suggest appropriate labels
Output: Labels + confidence scores
Speed: 1-2 seconds

​
Models

Currently supports:
  • text-embedding-004 - Embeddings (default, recommended)
Other models can be added in future versions.

​
Configuration

embedding:
  provider: "gemini"
  api_key: "${GEMINI_API_KEY}"
  model: "text-embedding-004"
  dimensions: 768
  batch_size: 100

​
API Quotas

Free Tier:
  • Embeddings: 50 requests/minute
  • LLM Calls: 15 requests/minute
  • Generous monthly limits
Paid:
  • Pay-as-you-go
  • Higher quotas available

​
Error Handling

Graceful degradation if Gemini unavailable: 
If embeddings fail:
  → Step skipped
  → No similarity search
  → Continue with other analysis

If LLM fails:
  → That analysis step skipped
  → Other steps continue
  → Error logged

​
Performance

Typical latencies:
  • Embedding request: 200-500ms
  • LLM analysis: 2-5 seconds
  • Batch embedding: 500ms-2s
Bottleneck is usually external API calls.

​
Cost Estimation

For 1,000 issues:
  • Embeddings: ~$0.01-0.05 (bulk indexing)
  • LLM analysis: $0.10-0.50 per issue (depends on features)
  • Typical total: $100-500/month for active repo
Optimize by:
  • Disabling unnecessary features
  • Using similarity-only workflow
  • Archiving old issues
  • Batching operations

​
Next Steps

Gemini Configuration

Setup Gemini for Simili Bot

Integrations

Other integrations