DeepSeek V3.2 pricing across providers
Across tracked provider offerings, input runs $0.186–$3.00 per million tokens and output $0.280–$4.50 per million tokens. The cheapest tracked provider offering for a typical agentic mix (2.5% input, 97% cached input, 0.5% output) is Avian at $0.019 per million tokens.
Prices are USD per million tokens and reflect each provider offering — they are not a single intrinsic model price. 9 of 12 offerings publish a cache-read rate; 5 of 12 publish a 30-minute endpoint uptime figure.
Open the calculator filtered to DeepSeek V3.2
Is this the right model for your workload? Compare open-weight picks and providers for your use case →
Daily price history
Loading price history…
Daily snapshots are retained for up to 90 days; history begins with the first recorded snapshot.
Provider offerings for DeepSeek V3.2
Ranked by effective cost at a typical agentic mix. Quantized and tier variants stay separate rows.
| Provider | Quant | Input $/M | Output $/M | Cache $/M | Blended $/M | Uptime (30m) |
|---|---|---|---|---|---|---|
| Avian | — | $0.230 | $0.330 | $0.012 | $0.019 | — |
| GMI Cloud promo | fp8 | $0.209 | $0.310 | $0.022 | $0.028 | 100.00% |
| Baidu | fp8 | $0.280 | $0.420 | $0.028 | $0.036 | 100.00% |
| Bytedance | — | $0.280 | $0.420 | $0.056 | $0.063 | — |
| Digitalocean | unknown | $0.300 | $0.960 | $0.090 | $0.100 | 100.00% |
| DeepInfra | — | $0.260 | $0.380 | $0.130 | $0.135 | — |
| Atlascloud | fp8 | $0.260 | $0.380 | $0.130 | $0.135 | 100.00% |
| SiliconFlow | fp8 | $0.259 | $0.800 | $0.140 | $0.146 | 100.00% |
| Singularity | — | $0.186 | $0.280 | — | $0.186 | — |
| Phala | unknown | $1.00 | $1.00 | $0.500 | $0.515 | — |
| SambaNova | — | $3.00 | $4.50 | — | $3.01 | — |
| Mara | unknown | $3.00 | $4.50 | — | $3.01 | — |
Provider-offering pricing refreshed 2026-10-29. Uptime is the 30-minute endpoint metric where a provider publishes it. Verify rates on the provider's official pricing page before committing spend.