# AI Model Performance Dashboard

Multi-provider AI performance dashboard with context occupancy tracking, truncation warnings, TTFT latency, tokens/min rate, and model comparison metrics.

---


## Metadata

**Title:** AI Model Performance Dashboard
**Category:** statuslines
**Author:** JSONbored
**Added:** October 2025
**Tags:** dashboard, performance, multi-model, metrics, occupancy, latency, production
**URL:** https://claudepro.directory/statuslines/ai-model-performance-dashboard

## Overview

Multi-provider AI performance dashboard with context occupancy tracking, truncation warnings, TTFT latency, tokens/min rate, and model comparison metrics.

## Content

#!/usr/bin/env bash
AI MODEL PERFORMANCE DASHBOARD FOR CLAUDE CODE
DISPLAYS: OCCUPANCY % | TRUNCATION | TTFT | TOKENS/MIN | MODEL LIMITS
READ JSON FROM STDIN
read -r input
EXTRACT VALUES
model=$(echo "$input" | jq -r '.model // "unknown"')
prompt_tokens=$(echo "$input" | jq -r '.session.promptTokens // 0')
completion_tokens=$(echo "$input" | jq -r '.session.completionTokens // 0')
total_tokens=$(echo "$input" | jq -r '.session.totalTokens // 0')
session_start=$(echo "$input" | jq -r '.session.startTime // ""')
MODEL CONTEXT LIMITS ( VERIFIED)
case "$model" in
 "claude-sonnet-4"|"claude-4")
 context_limit=
 model_display="Claude Sonnet 4"
 ;;
 "claude-3.5"|"claude-sonnet-3.5")
 context_limit=
 model_display="Claude 3.5 Sonnet"
 ;;
 "gpt-4.1"|"gpt-4-turbo")
 context_limit=
 model_display="GPT-4.1 Turbo"
 ;;
 "gpt-4o")
 context_limit=
 model_display="GPT-4o"
 ;;
 "gemini-1.5-pro")
 context_limit=
 model_display="Gemini 1.5 Pro"
 ;;
 "gemini-2")
 context_limit=
 model_display="Gemini 2.x"
 ;;
 "grok-3")
 context_limit=
 model_display="Grok 3"
 ;;
 "grok-4")
 context_limit=
 model_display="Grok 4"
 ;;
 "llama-4")
 context_limit=
 model_display="Llama 4 Scout"
 ;;
 *)
 context_limit=
 model_display="$model"
 ;;
esac
CALCULATE OCCUPANCY PERCENTAGE
if [ $context_limit -gt 0 ]; then
 occupancy=$((prompttokens * / contextlimit))
else
 occupancy=0
fi
OCCUPANCY COLOR CODING
if [ $occupancy -lt 50 ]; then
 OCC_COLOR="\[38;5;46m" # Green: 80%
fi
TRUNCATION WARNING (MODELS FAIL BEFORE ADVERTISED LIMITS)
reliablelimit=$((contextlimit * 65 / )) # 65% of claimed limit
if [ $prompttokens -gt $reliablelimit ]; then
 truncation="⚠ TRUNCATION RISK"
 TRUNC_COLOR="\[38;5;196m"
else
 truncation="✓ Safe"
 TRUNC_COLOR="\[38;5;46m"
fi
CALCULATE TOKENS PER MINUTE
if [ -n "$session_start" ]; then
 current_time=$(date +%s)
 startepoch=$(date -j -f "%Y-%m-%dT%H:%M:%SZ" "$sessionstart" +%s 2>/dev/null || echo "$current_time")
 elapsedseconds=$((currenttime - start_epoch))
 if [ $elapsed_seconds -gt 0 ]; then
 tokenspermin=$((totaltokens * 60 / elapsedseconds))
 else
 tokenspermin=0
 fi
else
 tokenspermin=0
fi
FORMAT NUMBERS WITH COMMAS
formatNumber() {
 printf "%'d" "$1" 2>/dev/null || echo "$1"
}
TTFT SIMULATION (TIME TO FIRST TOKEN - WOULD NEED ACTUAL TIMING)
FOR DEMO PURPOSES, ESTIMATE BASED ON TOKENS
if [ $prompt_tokens -gt ]; then
 ttft="~2.5s"
 TTFT_COLOR="\[38;5;226m"
elif [ $prompt_tokens -gt ]; then
 ttft="~1.2s"
 TTFT_COLOR="\[38;5;46m"
else
 ttft="~0.8s"
 TTFT_COLOR="\[38;5;46m"
fi
RESET="\[0m"
BOLD="\[1m"
BUILD DASHBOARD (MULTI-LINE FOR COMPREHENSIVE VIEW)
echo -e "${BOLD}📊 ${model_display}${RESET}"
echo -e "${OCCCOLOR}Occupancy: ${occupancy}%${RESET} ($(formatNumber $prompttokens)/$(formatNumber $contextlimit) tokens) | ${TRUNCCOLOR}${truncation}${RESET}"
echo -e "${TTFTCOLOR}TTFT: ${ttft}${RESET} | Rate: ${tokenspermin} tok/min | Total: $(formatNumber $totaltokens)"
KEY FEATURES
? Multi-provider model support (Claude, GPT, Gemini, Grok, Llama)
? Context occupancy percentage with color-coded warnings
? Truncation risk alerts based on real-world model reliability limits (65%)
? Time to First Token (TTFT) estimation for latency monitoring
? Tokens per minute rate calculation for throughput tracking
? Formatted number display with thousands separators
? verified context limits for all major models
? Multi-line dashboard for comprehensive metrics at a glance
CONFIGURATION
Format: bash
Refresh Interval: 2000ms
Position: left
Color Scheme: traffic-light-metrics
USE CASES
? Production workflows requiring performance monitoring
? Multi-model comparison and optimization
? Preventing context truncation in long conversations
? Tracking token consumption rates for cost management
? Identifying latency issues before they impact workflow
? SLA monitoring for enterprise AI deployments
? Performance regression detection across model versions
TROUBLESHOOTING
1) Occupancy percentage always showing 0% or incorrect values
 Solution: Verify session.promptTokens exists in JSON: jq .session.promptTokens. Check context_limit set correctly for model. Test calculation: echo $(( * / )) (should be 10). Ensure integer division working.
2) Model not recognized, showing default context_limit 
 Solution: Check model string extraction: jq -r .model. Verify case statement matches model name pattern. Add custom model: Add case for "your-model") context_limit=X ;; before default case.
3) Tokens per minute showing 0 or extremely high numbers
 Solution: Verify session.startTime format: jq -r .session.startTime. Check date parsing works: date -j -f '%Y-%m-%dT%H:%M:%SZ' '-23T10:00:00Z' +%s. Linux users use -d instead of -j. Ensure elapsed_seconds > 0.
4) Number formatting with commas not working (showing raw numbers)
 Solution: Check printf thousands separator support: printf "%'d" . If unsupported, formatNumber falls back to raw echo. Enable locale: export LCNUMERIC=enUS.UTF-8. Test: locale | grep LC_NUMERIC.
5) Truncation warning appearing too early or too late
 Solution: Adjust reliablelimit calculation (currently 65% of contextlimit). Research shows models fail % before claimed limits. Increase for conservative: reliablelimit=$((contextlimit * 50 / )). Decrease for aggressive: 75%.
TECHNICAL DETAILS
Documentation: https://epoch.ai/data-insights/context-windows
PREVIEW
📊 Claude Sonnet 4
Occupancy: 42% (,/1,, tokens) | ✓ Safe
TTFT: ~1.2s | Rate: 1, tok/min | Total: ,


---

Source: Claude Pro Directory
Website: https://claudepro.directory
URL: https://claudepro.directory/statuslines/ai-model-performance-dashboard

This content is optimized for Large Language Models (LLMs).
For full formatting and interactive features, visit the website.