The board shows this model's cheapest measured offer ($0.544/1M); the smaller figure under it is the mean of the 30 offers listed below ($1.69/1M). Add those up and divide — it matches. 24h token volume (426B) is sourced from OpenRouter public rankings — today's only public source of absolute per-model token counts (Vercel AI Gateway publishes share-of-traffic percentages); as more sources come online, the per-router breakdown will sum to this total here. Cross-check — Vercel AI Gateway: 4.61% of gateway token volume, —% of spend (2026-08-04; share-of-traffic, daily, data CC BY 4.0).
| Provider | Router / Source | $ in /1M | $ out /1M | TTFT | TPS | Uptime | Evidence |
|---|---|---|---|---|---|---|---|
| deepseek | Pricebook (LiteLLM) | $0.435 | $0.870 | — | — | — | Official list |
| tencent | Pricebook (LiteLLM) | $0.435 | $0.870 | — | — | — | Official list |
| requesty | Requesty | $0.435 | $0.870 | — | — | — | API-measured |
| DeepSeek | OpenRouter | $0.435 | $0.870 | 1317ms | 32 | 100.0% | API-measured |
| StreamLake (fp8) | OpenRouter | $0.652 | $1.30 | 2583ms | 43 | 97.4% | API-measured |
| GMICloud (fp8) | OpenRouter | $0.661 | $1.32 | 2762ms | 51 | 97.7% | API-measured |
| Ionstream (fp4) | OpenRouter | $1.13 | $2.26 | 4829ms | 6 | 93.9% | API-measured |
| Novita (fp8) | OpenRouter | $1.17 | $2.34 | 1680ms | 62 | 99.8% | API-measured |
| Baidu (fp8) | OpenRouter | $1.23 | $2.47 | 718ms | 64 | 99.6% | API-measured |
| deepinfra | HuggingFace Router | $1.30 | $2.60 | 398ms | 38 | — | API-measured |
| DeepInfra (fp4) | OpenRouter | $1.30 | $2.60 | 1100ms | 16 | 98.5% | API-measured |
| DigitalOcean | OpenRouter | $1.39 | $2.78 | 2389ms | 8 | 99.3% | API-measured |
| Alibaba (fp8) | OpenRouter | $1.42 | $2.83 | 1824ms | 59 | 99.8% | API-measured |
| 0g-teetls | 0G Router | $1.45 | $2.90 | — | — | — | API-measured |
| SiliconFlow (fp8) | OpenRouter | $1.50 | $3.13 | 1639ms | 50 | 99.9% | API-measured |
| novita | HuggingFace Router | $1.60 | $3.20 | 544ms | 28 | — | API-measured |
| novita-ai | Novita AI | $1.60 | $3.20 | — | — | — | API-measured |
| burncloud | burncloud | $1.64 | $3.28 | — | — | — | API-measured |
| Venice | OpenRouter | $1.65 | $3.30 | 1230ms | 25 | 97.6% | API-measured |
| AtlasCloud (fp4) | OpenRouter | $1.68 | $3.38 | 2290ms | 49 | 99.3% | API-measured |
| azure_ai | Pricebook (LiteLLM) | $1.74 | $3.48 | — | — | — | Official list |
| fireworks_ai | Pricebook (LiteLLM) | $1.74 | $3.48 | — | — | — | Official list |
| together | HuggingFace Router | $1.74 | $3.48 | 616ms | 47 | — | API-measured |
| fireworks-ai | HuggingFace Router | $1.74 | $3.48 | 968ms | 60 | — | API-measured |
| BaseTen (fp4) | OpenRouter | $1.74 | $3.48 | 545ms | 107 | 98.5% | API-measured |
| Parasail (fp8) | OpenRouter | $1.74 | $3.48 | 1967ms | 42 | 99.7% | API-measured |
| Cloudflare | OpenRouter | $1.74 | $3.48 | 2136ms | 42 | 99.4% | API-measured |
| Together | OpenRouter | $1.74 | $3.48 | 522ms | 60 | 99.1% | API-measured |
| CoreWeave (fp8) | OpenRouter | $1.74 | $3.48 | 1680ms | 112 | 68.2% | API-measured |
| Fireworks | OpenRouter | $1.74 | $3.48 | 1579ms | 42 | — | API-measured |
Volume trend
Last 30 days: 359B → 426B (+19%); range 262B–597B tokens/day. Source: OpenRouter public rankings; refreshed hourly.
Data & method
Cheapest ranked by blended = (3×in + out) ÷ 4; performance columns show only when a source provides them, never estimated. “Official list” = LiteLLM open pricebook (MIT); “API-measured” = provider/aggregator public APIs. Machine-readable data: /api/v1/models/deepseek-v4-pro.json. Free reuse requires attribution to tkx.org.
About
Deepseek-v4-pro is an AI model, a machine-learning system designed for language understanding and generation tasks. It is served by multiple inference providers on AI model marketplaces under the organization tag "deepseek".
AI-generated summary from public information — is this yours?
FAQ
How much does deepseek-v4-pro cost?
Cheapest measured offer right now: $0.435/1M input, $0.87/1M output via deepseek — across 23 tracked providers, refreshed hourly.
Who serves deepseek-v4-pro?
23 providers currently serve it. The table above lists each with input/output price, TTFT, TPS, uptime and an evidence tier.
How heavily is deepseek-v4-pro used?
deepseek-v4-pro routed 426B tokens in the last 24h. The daily bars above show routed volume over recent days. Source: OpenRouter public rankings — one of the few routers publishing per-model usage; TKX refreshes hourly.