DeepSeek: DeepSeek V3.2
deepseek/deepseek-v3.2
Cheapest provider
$0.25 / 1M
Baidu
Fastest provider (p95)
—
No throughput data yet — populated as traffic accumulates
Intelligence (composite)
83.3
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 |
|---|---|---|---|---|---|---|---|
| Baidu | q8· fp8 | $0.2520 | $0.3780 | — | — | 99.68% | — |
| StreamLake | undisclosed | $0.2574 | $0.3861 | — | — | 99.35% | — |
| SiliconFlow | q8· fp8 | $0.2590 | $0.4200 | — | — | 99.69% | — |
| AtlasCloud | q8· fp8 | $0.2600 | $0.3800 | — | — | 96.21% | — |
| DeepInfra | q4· fp4 | $0.2600 | $0.3800 | — | — | 99.99% | — |
| Novita | q8· fp8 | $0.2690 | $0.4000 | — | — | 99.74% | — |
| Parasail | q8· fp8 | $0.2800 | $0.4500 | — | — | 95.93% | — |
| Alibaba | undisclosed | $0.3705 | $1.1115 | — | — | 98.81% | — |
| Friendli | undisclosed | $0.5000 | $1.5000 | — | — | 99.99% | — |
| undisclosed | $0.5600 | $1.6800 | — | — | 99.94% | — | |
| SambaNova | undisclosed | $3.0000 | $4.5000 | — | — | — | — |
“—” 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
85.0
broad reasoning
Code
79.0
pass@1 (HumanEval / LiveCodeBench)
AIME 2025
88.0
math accuracy
GPQA Diamond
80.0
hard reasoning
Source: Artificial Analysis (AA-aligned) via MiniMax-M2 table — DeepSeek-V3.2 (MMLU-Pro 85, GPQA-D 80, LiveCodeBench 79, AIME25 88); github.com/MiniMax-AI/MiniMax-M2
How Atlas routes DeepSeek: DeepSeek V3.2
- 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 DeepSeek: DeepSeek V3.2 that’s currently Baidu at $0.25/1M.
- Batch tier — async, biggest discount. Roadmapped to use provider batch APIs (OpenAI / Anthropic) where available and queued spot capacity for open-weight workloads.