A sample LLMIntel dashboard with a week of realistic traffic across three apps.
You’re viewing demo data — this is what LLMIntel looks like with a week of real traffic. Sign up and your own data appears minutes after your first instrumented call.
| Model | State | Retirement |
|---|---|---|
| meta.llama3-1-405b-instruct-v1:0 | retiring | in 21 days2026-10-24 |
Pricing known for 6 of 5 API-served models you run. 1 more is self-hosted, where cost comes from your own infrastructure rather than a per-token rate. Cheapest by input: gpt-5-nano-2025-08-07 at $0.05/1M.
Want cheaper/faster on-par alternatives for these models? See optimization →
Times in UTC · 13 months of history on your plan · sample data — send telemetry to see your own, range-filtered
Window: last 30 days.
Telemetry up to Oct 3, 17:04 UTC· 8 min ago· Catalog checked 42 min ago
These models are returning more output per input token than their 30-day baseline — the usual cause of silent cost creep. Check prompts, max_tokens, and whether a cheaper or more concise model would do.
| Model | Now | Baseline | Drift | Requests | Spend |
|---|---|---|---|---|---|
| claude-sonnet-4-5-20250929 | 0.3× | 0.2× | +38% | 101,040 | $620.86 |
Dollars flowing through models that retire within 90 days. Migrate before the provider 4xxs you — see recommended replacements on each model page.
| Model | Retirement | Spend |
|---|---|---|
| meta.llama3-1-405b-instruct-v1:0 | in 21 days2026-10-24 | $212.40 |
| Environment › Application | Requests | Tokens | Blended $/1M | Cost/req | Spend | Share |
|---|---|---|---|---|---|---|
| prod | 379,000 | 86M | $13.16 | $0.0030 | $1,131.66 | 88% |
| Checkout Assistant | 261,000 | 58M | $12.77 | $0.0028 | $742.11 | 58% |
| Support Copilot | 118,000 | 28M | $13.96 | $0.0033 | $389.55 | 30% |
| staging | 33,000 | 10M | $14.65 | $0.0046 | $152.40 | 12% |
| Unassignedno app mapping | 33,000 | 10M | $14.65 | $0.0046 | $152.40 | 12% |
app tag your agent sends or the API key's app mapping; calls with neither show as Unassigned (map keys under Applications or send an app tag). Unattributed is usage we couldn't tie to a specific key. Every dollar lands in exactly one bucket, and the tree follows the active filters (including the tag filter above). Share is relative to the total spend in view.| feature | Requests | Tokens | Blended $/1M | Cost/req | Spend | Share |
|---|---|---|---|---|---|---|
| checkout | 214,000 | 48M | $12.51 | $0.0028 | $604.32 | 47% |
| search | 121,000 | 28M | $14.54 | $0.0033 | $401.18 | 31% |
| summarize | 54,000 | 13M | $14.68 | $0.0035 | $187.96 | 15% |
| Untaggedno feature tag | 23,000 | 7.7M | $11.77 | $0.0039 | $90.60 | 7% |
| team | Requests | Tokens | Blended $/1M | Cost/req | Spend | Share |
|---|---|---|---|---|---|---|
| growth | 268,000 | 61M | $12.75 | $0.0029 | $771.24 | 60% |
| platform | 144,000 | 33M | $13.70 | $0.0031 | $452.22 | 35% |
| Untaggedno team tag | 0 | 4.3M | $14.09 | — | $60.60 | 5% |
| Model | Provider | Requests | Input | Output | Blended $/1M | Cost/req | Tokens/req | Out:in | Spend |
|---|---|---|---|---|---|---|---|---|---|
| claude-sonnet-4-5-20250929 | Anthropic | 168,400 | 31M 14M cached | 8.5M | $15.64 | $0.0037 | 236 | 0.3× | $620.86 |
| gpt-5-2025-08-07 | OpenAI | 96,800 | 22M 9.0M cached | 6.9M | $9.98 | $0.0030 | 303 | 0.3× | $292.40 |
| gpt-5-mini-2025-08-07 | OpenAI | 74,300 | 12M 3.6M cached | 3.4M | $2.09 | $0.0004 | 209 | 0.3× | $32.40 |
| gpt-5-nano-2025-08-07 | OpenAI | 41,900 | 5.6M | 1.5M | $1.13 | $0.0002 | 169 | 0.3× | $8.00 |
| gpt-4o (2024-08-06) | Azure AI Foundry | 22,600 | 2.9M | 900k | $31.05 | $0.0052 | 168 | 0.3× | $118.00 |
| meta.llama3-1-405b-instruct-v1:0 | AWS Bedrock | 8,000 | 780k | 220k | $212.40 | $0.0266 | 125 | 0.3× | $212.40 |
$212/mo of your spend sits on models a provider is retiring, the first on 2026-10-24. Adopting every replacement below adds $71/mo.
Up to $1.8k/mo in projected savings at your last-30-day usage, if you adopt the switches below. Estimates reprice your own token mix — advisory, not a quote.