OpenAI: GPT-4o-mini
openai/gpt-4o-mini
Cheapest provider
$0.15 / 1M
Azure
Fastest provider (p95)
67 tok/s
OpenAI
Intelligence (estimated)
66.6
Family + generation + popularity + price tier
Per-provider performance
Latency / throughput / uptime measured across providers over the last 30 minutes of live traffic. Atlas’s router weighs these per call (with the eval-gate signal) when picking a variant for Standard and Batch tiers.
| Provider | Quant | Input $/1M | Output $/1M | Latency p50 / p95 | Throughput p50 / p95 | Uptime 30m | Success |
|---|---|---|---|---|---|---|---|
| Azure | full· unknown | $0.1500 | $0.6000 | 2568ms / 8421ms | 8 / 16 tok/s | 89.82% | 93.8% |
| Azure | full· unknown | $0.1500 | $0.6000 | 2568ms / 8421ms | 8 / 16 tok/s | 94.21% | 93.8% |
| OpenAI | full· unknown | $0.1500 | $0.6000 | 716ms / 4091ms | 28 / 67 tok/s | 99.41% | 99.4% |
“—” means live telemetry hasn’t accumulated enough recent traffic for that endpoint. “undisclosed” means the provider serves the model but doesn’t expose the quantization label (typically running fp8 / int8 internally).
Intelligence estimate
No public benchmark numbers indexed for this model yet, so the leaderboard score is derived from the catalogue data we sync: model family standing, generation, and live usage rank — applied identically to open- and closed-weight models. The heuristic ceiling sits below confirmed-benchmark frontier models so curated rankings stay clearly on top.
- ·openai mini family base
- ·top-16 on OpenRouter +1
Want a curated score? File a benchmark report and we’ll add it to the next sync.
How Atlas routes OpenAI: GPT-4o-mini
- Realtime tier — direct passthrough at the upstream’s native precision. Best for hard latency / quality guarantees.
- Standard tier — Atlas picks the cheapest provider variant whose quantization has stayed green on your operation’s eval gates. For OpenAI: GPT-4o-mini that’s currently Azure at $0.15/1M.
- Batch tier — async, biggest discount. Roadmapped to use provider batch APIs (OpenAI / Anthropic) where available and queued spot capacity for open-weight workloads.