DeepSeek API Providers
Who serves DeepSeek, at what price and speed.
DeepSeek releases its models with open weights, from V3 and R1 to the V4 Flash and Pro generation and V4.1 Flash.
DeepSeek served 14% of the 81.1T DeepSeek tokens that went through a large public LLM router over Nov 2 to Nov 8, 2026. DeepSeek's own API served 14%; 46 providers list the family.
Who serves DeepSeek tokens
| Provider | Tokens, 7d | Share |
|---|---|---|
| DeepSeek (lab) | 11.5T | 14.3% |
| DeepInfra | 10.2T | 12.7% |
| Wafer | 8.93T | 11.1% |
| Together | 7.62T | 9.5% |
| Relace | 7.51T | 9.3% |
| inference.net | 4.80T | 6.0% |
| Morph | 4.34T | 5.4% |
| DigitalOcean | 4.12T | 5.1% |
| Open Inference | 2.02T | 2.5% |
| CoreWeave | 1.67T | 2.1% |
| Fireworks | 1.65T | 2.1% |
| StreamLake | 1.62T | 2.0% |
| Crusoe | 1.35T | 1.7% |
| Baidu Qianfan | 1.31T | 1.6% |
| Alibaba Cloud Int. | 1.05T | 1.3% |
| 32 others | 10.8T | 13.4% |
All DeepSeek models combined, Nov 2 to Nov 8, 2026.
DeepSeek V4.1 Flash 1017: prices and speed by provider
Output prices run from $0.15 to $2.40 per million tokens across 40 providers, median $0.675.
| Provider | Share | Input $/M | Output $/M | Cache read $/M | Tokens/s | Latency | Uptime 3d | Context |
|---|---|---|---|---|---|---|---|---|
| DeepSeek | 20.3% | $0.15 | $0.60 | $0.0030 | 114 | 1.30s | 98.9% | 1.05M |
| DeepInfra | 13.4% | $0.078 | $0.235 | $0.0024 | 47 | 2.84s | 100.0% | 1.05M |
| Together | 9.9% | $0.30 | $1.20 | $0.0060 | 188 | 0.56s | 100.0% | 1.05M |
| Wafer | 9.1% | $0.038 | $0.50 | $0.01 | 135 | 0.42s | 100.0% | 1.05M |
| Relace | 9.0% | $0.0060 | $0.50 | $0.0050 | 79 | 0.82s | 100.0% | 1.05M |
| inference.net | 7.3% | $0.08 | $0.55 | $0.0080 | 113 | 0.26s | 100.0% | 1.04M |
| Morph | 5.5% | $0.022 | $0.40 | $0.0040 | 84 | 0.98s | 99.7% | 1.05M |
| DigitalOcean | 4.4% | $0.18 | $0.72 | $0.0060 | 151 | 0.88s | 100.0% | 1.05M |
| Open Inference | 2.2% | $0.024 | $0.70 | $0.0050 | 116 | 0.89s | 100.0% | 1.05M |
| Fireworks | 1.8% | $0.21 | $0.84 | $0.0042 | 68 | 2.67s | 96.8% | 1.05M |
| Particle AI | 1.3% | $0.025 | $0.25 | $0.0015 | 69 | 0.78s | 100.0% | 1.05M |
| Ionstream | 1.2% | $0.083 | $0.247 | $0.0052 | 97 | 1.31s | - | 1.05M |
| Alibaba Cloud Int. | 1.2% | $0.075 | $0.30 | $0.0075 | 93 | 2.18s | 100.0% | 1M |
| Crusoe | 1.1% | $0.29 | $1.20 | $0.0070 | 175 | 0.86s | 100.0% | 1.05M |
| StepFun | 1.1% | $0.135 | $0.54 | $0.0027 | 131 | 1.04s | 100.0% | 1M |
| StreamLake | 1.0% | $0.111 | $0.444 | $0.0022 | 62 | 1.32s | 100.0% | 1.02M |
| Cloudflare | 1.0% | $0.10 | $0.60 | $0.0030 | 88 | 1.98s | 100.0% | 1.05M |
| Cohere | 1.0% | $0.195 | $0.66 | $0.0060 | 217 | 0.36s | 100.0% | 1.05M |
| Parasail | 0.9% | $0.30 | $1.20 | $0.0060 | 126 | 0.76s | 100.0% | 1.05M |
| Modal | 0.9% | $0.25 | $1.00 | $0.0050 | 125 | 0.83s | 100.0% | 1.05M |
| Prime Intellect | 0.8% | $0.044 | $0.70 | $0.0024 | 65 | 1.17s | 100.0% | 1.05M |
| NovitaAI | 0.8% | $0.165 | $0.66 | $0.0033 | 71 | 1.65s | 100.0% | 1.05M |
| Baseten | 0.7% | $0.30 | $1.20 | $0.0070 | 145 | 0.48s | 99.6% | 1.05M |
| SiliconFlow | 0.6% | $0.30 | $0.55 | $0.0060 | 81 | 0.82s | 100.0% | 1.05M |
| ModelRun [by Modular] | 0.6% | $0.17 | $0.67 | $0.0060 | 117 | 0.94s | 100.0% | 1.05M |
| Sail Research | 0.5% | $0.101 | $0.15 | $0.0067 | - | - | 27.5% | 1.05M |
| Decart | 0.4% | $0.123 | $0.492 | $0.0025 | 101 | 1.69s | 99.8% | 1.05M |
| Baidu Qianfan | 0.4% | $0.28 | $1.13 | $0.0060 | 76 | 2.89s | 99.7% | 1.05M |
| CoreWeave | 0.3% | $0.30 | $1.20 | $0.0060 | 211 | 0.47s | 97.9% | 1.05M |
| Mistral | 0.3% | $0.30 | $1.20 | $0.0070 | 53 | 4.69s | 99.8% | 1.05M |
| DekaLLM | 0.2% | $0.18 | $0.68 | $0.0060 | 76 | 1.85s | 99.9% | 1.05M |
| AtlasCloud | 0.2% | $0.195 | $0.78 | $0.0039 | 74 | 1.97s | 100.0% | 1.05M |
| Makora | 0.1% | $0.15 | $1.20 | $0.0060 | 145 | 0.71s | 100.0% | 1.05M |
| GMICloud | 0.1% | $0.18 | $0.72 | $0.0036 | 119 | 1.67s | 100.0% | 1.05M |
| Reka AI | <0.1% | $0.16 | $1.10 | $0.0080 | 97 | 1.24s | 99.8% | 1.05M |
| Nebius Token Factory | <0.1% | $0.30 | $1.20 | $0.0060 | 96 | 1.23s | 100.0% | 1.05M |
| Venice | <0.1% | $0.30 | $1.20 | $0.0075 | 102 | 1.60s | 100.0% | 1M |
| Phala | <0.1% | $0.20 | $1.20 | $0.0060 | 73 | 1.59s | 100.0% | 1.05M |
| PrimaLabs | <0.1% | $0.24 | $0.96 | $0.0048 | 137 | 1.99s | 98.7% | 1.05M |
| NEAR AI | <0.1% | $0.30 | $1.20 | $0.0060 | 42 | 0.55s | 100.0% | 1.05M |
List prices as of November 8, 2026, cheapest endpoint per provider. Tokens/s and latency are medians over 30 minutes.
DeepSeek V4.1 Flash: prices and speed by provider
Output prices run from $0.30 to $1.80 per million tokens across 30 providers, median $0.84.
| Provider | Share | Input $/M | Output $/M | Cache read $/M | Tokens/s | Latency | Uptime 3d | Context |
|---|---|---|---|---|---|---|---|---|
| Wafer | 15.8% | $0.051 | $0.567 | $0.018 | 114 | 0.58s | 100.0% | 1.05M |
| Together | 11.8% | $0.30 | $1.20 | $0.0060 | 211 | 0.26s | 100.0% | 1.05M |
| DeepInfra | 11.4% | $0.112 | $0.336 | $0.0034 | - | - | 41.5% | 1.05M |
| Relace | 8.6% | $0.0060 | $0.50 | $0.0050 | 74 | 0.94s | 99.5% | 1.05M |
| Morph | 6.9% | $0.022 | $0.40 | $0.0040 | 58 | 0.92s | 99.8% | 1.05M |
| CoreWeave | 6.8% | $0.13 | $0.65 | $0.0020 | 90 | 0.75s | 100.0% | 1.05M |
| DigitalOcean | 6.7% | $0.225 | $0.90 | $0.0060 | 116 | 0.89s | 100.0% | 1.05M |
| Open Inference | 4.4% | $0.024 | $0.70 | $0.0050 | 112 | 0.87s | 100.0% | 1.05M |
| inference.net | 4.0% | $0.103 | $0.60 | $0.0080 | 81 | 0.41s | 100.0% | 1.04M |
| Fireworks | 3.5% | $0.21 | $0.84 | $0.0042 | 106 | 0.94s | 99.8% | 1.05M |
| Baseten | 3.3% | $0.30 | $1.20 | $0.0070 | 151 | 0.94s | 99.4% | 1.05M |
| Baidu Qianfan | 2.8% | $0.30 | $1.20 | $0.0060 | 54 | 1.21s | 100.0% | 1.05M |
| Crusoe | 2.8% | $0.29 | $1.20 | $0.0070 | 119 | 0.63s | 100.0% | 1.05M |
| StreamLake | 2.7% | $0.147 | $0.588 | $0.0029 | 86 | 0.91s | 100.0% | 1.02M |
| Alibaba Cloud Int. | 2.1% | $0.075 | $0.30 | $0.0075 | 107 | 0.49s | 100.0% | 1M |
| Prime Intellect | 1.9% | $0.039 | $1.20 | $0.0012 | 98 | 0.69s | 99.6% | 1.05M |
| Parasail | 1.2% | $0.30 | $1.20 | $0.0060 | 177 | 0.32s | 100.0% | 1.05M |
| Venice | 0.9% | $0.30 | $1.20 | $0.0075 | 130 | 1.10s | 99.3% | 1M |
| Cohere | 0.6% | $0.195 | $0.66 | $0.0030 | 122 | 0.50s | 99.9% | 1.05M |
| NovitaAI | 0.5% | $0.195 | $0.78 | $0.0039 | 95 | 0.95s | 100.0% | 1.05M |
| Makora | 0.4% | $0.26 | $1.17 | $0.0058 | 88 | 0.53s | 100.0% | 1.05M |
| AtlasCloud | 0.2% | $0.135 | $0.54 | $0.013 | 97 | 0.63s | 100.0% | 1.05M |
| Reka AI | 0.2% | $0.12 | $0.96 | $0.0080 | 66 | 0.62s | 99.7% | 1.05M |
| GMICloud | 0.1% | $0.18 | $0.72 | $0.0036 | 80 | 1.44s | 100.0% | 1.05M |
| Nebius Token Factory | 0.1% | $0.30 | $1.20 | $0.01 | 71 | 0.77s | 99.9% | 1.05M |
| SiliconFlow | <0.1% | $0.30 | $1.20 | $0.0060 | 52 | 1.38s | 99.9% | 1.05M |
| Modal | <0.1% | $0.30 | $1.20 | $0.03 | 132 | 0.46s | 100.0% | 1.05M |
| Phala | <0.1% | $0.21 | $0.84 | $0.0042 | 91 | 0.96s | 99.9% | 1.05M |
| Mancer | <0.1% | $0.125 | $0.50 | $0.05 | 43 | 1.00s | 99.9% | 1.05M |
| Google Vertex | - | $0.30 | $1.20 | $0.03 | 52 | 1.17s | 98.1% | 1.05M |
List prices as of November 8, 2026, cheapest endpoint per provider. Tokens/s and latency are medians over 30 minutes.
All DeepSeek models
| Model | Tokens, 7d | Providers | Output $/M | Context | Released |
|---|---|---|---|---|---|
| DeepSeek V4.1 Flash 1017 | 56.0T | 40 | $0.15 to $2.40 | 1.05M | 2026-10-17 |
| DeepSeek V4.1 Flash | 17.6T | 30 | $0.30 to $1.80 | 1.05M | 2026-09-10 |
| DeepSeek V4 Flash 0731 | 3.29T | 26 | $0.14 to $1.32 | 1.05M | 2026-07-31 |
| DeepSeek V4 Flash 0423 | 2.28T | 14 | $0.165 to $0.56 | 1.05M | 2026-04-24 |
| DeepSeek V4 Pro 0813 | 774B | 21 | $1.74 to $8.00 | 1.05M | 2026-08-12 |
| DeepSeek V4 Pro 0423 | 569B | 13 | $1.50 to $3.83 | 1.05M | 2026-04-24 |
| DeepSeek V3.2 | 286B | 7 | $0.31 to $1.00 | 164K | 2025-12-01 |
| DeepSeek V4 Flash Vision Exp | 69B | 2 | $0.647 to $1.32 | 1.05M | 2026-08-21 |
| DeepSeek V3.1 | 51B | 4 | $0.95 to $1.50 | 164K | 2025-08-21 |
| DeepSeek V3 0324 | 21B | 2 | $1.00 to $1.14 | 164K | 2025-03-24 |
| DeepSeek V3 | 19B | 2 | $0.712 to $1.03 | 164K | 2024-12-26 |
| DeepSeek V3.2 Exp | 12B | 1 | $0.41 | 164K | 2025-09-29 |
| R1 0528 | 11B | 3 | $2.15 to $2.29 | 164K | 2025-05-28 |
| DeepSeek V3.1 Terminus | 5B | 1 | $1.00 | 164K | 2025-09-22 |
Finance your GPUs
Serving DeepSeek on your own GPUs? Tell us the cluster, the models you serve and the contracts behind them, and we will share it with partner funders that fit.
Prefer email? hello@amcompute.com