Meta: Llama 4 Maverick
meta-llama/llama-4-maverick
Cheapest provider
$0.15 / 1M
DeepInfra
Fastest provider (p95)
—
No throughput data yet — populated as traffic accumulates
Intelligence (composite)
64.3
MMLU-Pro · HumanEval · math · GPQA
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 |
|---|---|---|---|---|---|---|---|
| DeepInfra | q8· fp8 | $0.1500 | $0.6000 | — | — | 99.95% | — |
| Novita | q8· fp8 | $0.2700 | $0.8500 | — | — | 100.00% | — |
| Parasail | q8· fp8 | $0.3500 | $1.0000 | — | — | 100.00% | — |
| SambaNova | undisclosed | $0.6300 | $1.8000 | — | — | 92.61% | — |
“—” 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 breakdown
Composite score is a weighted average of public benchmarks (30% MMLU-Pro, 25% code pass@1, 25% math, 20% GPQA). Numbers come from model cards and the Artificial Analysis intelligence harness; missing components are renormalised over what’s present.
MMLU-Pro
80.5
broad reasoning
Code
43.4
pass@1 (HumanEval / LiveCodeBench)
MATH
61.2
math accuracy
GPQA Diamond
69.8
hard reasoning
Source: Meta Llama 4 model card — Maverick instruct (MMLU-Pro 80.5, GPQA-Diamond 69.8, LiveCodeBench 43.4, MATH 61.2); llama.com/models/llama-4
How Atlas routes Meta: Llama 4 Maverick
- 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 Meta: Llama 4 Maverick that’s currently DeepInfra 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.