is now an ernd partner: 10% back on inference On this page
Cheapest and fastest
- Cheapest
- Particle AI (US, Flex)
- $0.0813 per 1M tokens blended
- $0.025 input, $0.25 output
- Fastest
- Baseten (Fast)
- 545 tokens per second
- $1.05 blended, 0.25 s to first token
Every provider
| Provider | Input /1M | Output /1M | Blended 3:1 | Cache read /1M | Context | Quantization | Latency | Throughput | With ernd card |
|---|---|---|---|---|---|---|---|---|---|
| Particle AI (US, Flex) | $0.025 | $0.25 | $0.0813 | $0.0015 | 1.05M | Not listed | 0.92 s | 110 tok/s | – |
| Sail Research (US) Down when checked | $0.101 | $0.15 | $0.113 | $0.0067 | 1.05M | FP4 | 0.71 s | 72 tok/s | – |
| Morph | $0.022 | $0.40 | $0.117 | $0.004 | 1.05M | FP8 | 0.70 s | 77 tok/s | – |
| DeepInfra (Flex) | $0.0784 | $0.235 | $0.118 | $0.0024 | 1.05M | FP8 | 3.00 s | 48 tok/s | – |
| Relace | $0.006 | $0.50 | $0.13 | $0.005 | 1.05M | FP4 | 0.68 s | 116 tok/s | – |
| Alibaba Cloud Int. | $0.075 | $0.30 | $0.131 | $0.0075 | 1M | Not listed | 0.86 s | 116 tok/s | – |
| DeepInfra | $0.098 | $0.294 | $0.147 | $0.0029 | 1.05M | FP8 | 1.18 s | 58 tok/s | – |
| Wafer (US) ernd partner | $0.038 | $0.50 | $0.153 | $0.01 | 1.05M | MXFP4 | 0.38 s | 167 tok/s | $0.138after 10% cash back |
| Particle AI (US) | $0.05 | $0.50 | $0.163 | $0.003 | 1.05M | Not listed | 1.05 s | 98 tok/s | – |
| Sail Research Down when checked | $0.151 | $0.225 | $0.17 | $0.0101 | 1.05M | FP4 | 0.95 s | 79 tok/s | – |
| Baidu Qianfan | $0.0977 | $0.394 | $0.172 | $0.0021 | 1.05M | FP8 | 1.19 s | 125 tok/s | – |
| Wafer ernd partner | $0.066 | $0.50 | $0.174 | $0.0065 | 1.05M | MXFP4 | 0.37 s | 165 tok/s | $0.157after 10% cash back |
| StreamLake | $0.111 | $0.444 | $0.194 | $0.0022 | 1.02M | FP8 | 1.88 s | 69 tok/s | – |
| Open Inference | $0.03 | $0.70 | $0.197 | $0.006 | 1.05M | FP4 | 0.86 s | 97 tok/s | – |
| inference.net | $0.08 | $0.55 | $0.198 | $0.008 | 1.04M | FP8 | 0.36 s | 105 tok/s | – |
| inference.net (US) | $0.08 | $0.55 | $0.198 | $0.008 | 1.04M | FP8 | 0.33 s | 111 tok/s | – |
| Decart | $0.123 | $0.492 | $0.215 | $0.0025 | 1.05M | FP8 | 1.21 s | 100 tok/s | – |
| DeepInfra (Priority) | $0.147 | $0.441 | $0.22 | $0.0044 | 1.05M | FP8 | 1.50 s | 59 tok/s | – |
| Cloudflare | $0.10 | $0.60 | $0.225 | $0.003 | 1.05M | Not listed | 2.16 s | 90 tok/s | – |
| StepFun | $0.135 | $0.54 | $0.236 | $0.0027 | 1M | FP8 | 0.79 s | 149 tok/s | – |
| Ionstream | $0.165 | $0.495 | $0.247 | $0.0105 | 1.05M | Not listed | 0.90 s | 119 tok/s | – |
| DeepSeek | $0.15 | $0.60 | $0.262 | $0.003 | 1.05M | Not listed | 1.09 s | 91 tok/s | – |
| NovitaAI | $0.165 | $0.66 | $0.289 | $0.0033 | 1.05M | FP8 | 1.71 s | 82 tok/s | – |
| ModelRun [by Modular] | $0.17 | $0.67 | $0.295 | $0.006 | 1.05M | FP4 | 0.85 s | 126 tok/s | – |
| DekaLLM | $0.18 | $0.68 | $0.305 | $0.006 | 1.05M | Not listed | 1.28 s | 112 tok/s | – |
| Cohere | $0.195 | $0.66 | $0.311 | $0.006 | 1.05M | FP8 | 0.52 s | 174 tok/s | – |
| DigitalOcean | $0.18 | $0.72 | $0.315 | $0.006 | 1.05M | Not listed | 0.82 s | 159 tok/s | – |
| GMICloud | $0.18 | $0.72 | $0.315 | $0.0036 | 1.05M | FP8 | 1.31 s | 120 tok/s | – |
| AtlasCloud | $0.195 | $0.78 | $0.341 | $0.0039 | 1.05M | FP8 | 1.48 s | 95 tok/s | – |
| SiliconFlow | $0.30 | $0.55 | $0.362 | $0.006 | 1.05M | FP8 | 0.99 s | 138 tok/s | – |
| Fireworks Spot | $0.21 | $0.84 | $0.367 | $0.0042 | 1.05M | Not listed | 1.13 s | 121 tok/s | – |
| Prime Intellect | $0.109 | $1.20 | $0.382 | $0.006 | 1.05M | Not listed | 0.89 s | 140 tok/s | – |
| Reka AI | $0.16 | $1.10 | $0.395 | $0.008 | 1.05M | Not listed | 1.15 s | 93 tok/s | – |
| Makora | $0.15 | $1.20 | $0.412 | $0.006 | 1.05M | FP8 | 1.06 s | 147 tok/s | – |
| PrimaLabs | $0.24 | $0.96 | $0.42 | $0.0048 | 1.05M | Not listed | 0.92 s | 140 tok/s | – |
| Modal | $0.25 | $1.00 | $0.438 | $0.005 | 1.05M | Not listed | 0.62 s | 149 tok/s | – |
| Relace (Fast) | $0.30 | $0.90 | $0.45 | $0.008 | 1.05M | FP4 | 0.62 s | 226 tok/s | – |
| Phala | $0.20 | $1.20 | $0.45 | $0.006 | 1.05M | Not listed | 1.81 s | 83 tok/s | – |
| Crusoe | $0.29 | $1.20 | $0.517 | $0.007 | 1.05M | FP8 | 0.55 s | 178 tok/s | – |
| Baseten | $0.30 | $1.20 | $0.525 | $0.007 | 1.05M | Not listed | 0.54 s | 143 tok/s | – |
| CoreWeave | $0.30 | $1.20 | $0.525 | $0.006 | 1.05M | FP8 | 0.55 s | 170 tok/s | – |
| Fireworks | $0.30 | $1.20 | $0.525 | $0.006 | 1.05M | Not listed | 1.23 s | 126 tok/s | – |
| Mistral | $0.30 | $1.20 | $0.525 | $0.007 | 1.05M | FP8 | 1.06 s | 186 tok/s | – |
| NEAR AI | $0.30 | $1.20 | $0.525 | $0.006 | 1.05M | Not listed | 0.67 s | 43 tok/s | – |
| Nebius Token Factory | $0.30 | $1.20 | $0.525 | $0.006 | 1.05M | FP8 | 1.18 s | 118 tok/s | – |
| Parasail | $0.30 | $1.20 | $0.525 | $0.006 | 1.05M | FP8 | 0.62 s | 149 tok/s | – |
| Together | $0.30 | $1.20 | $0.525 | $0.006 | 1.05M | Not listed | 0.26 s | 240 tok/s | – |
| Venice | $0.30 | $1.20 | $0.525 | $0.0075 | 1M | FP8 | 1.15 s | 120 tok/s | – |
| Fireworks (US) | $0.45 | $1.80 | $0.788 | $0.009 | 1.05M | Not listed | 0.93 s | 162 tok/s | – |
| Baseten (Fast) | $0.60 | $2.40 | $1.05 | $0.014 | 1.05M | Not listed | 0.25 s | 545 tok/s | – |
| Fireworks (Ultrafast) | $2.40 | $9.60 | $4.20 | $0.048 | 1.05M | Not listed | – | 213 tok/s | – |
Blended is three parts input price to one part output, per million tokens. Latency is time to first token and throughput is output tokens per second, both averaged from OpenRouter's daily figures for the last 7 days. Cheapest and fastest only count listings that were up when checked.
Wafer is an ernd partner: 10% back on Wafer inference with the ernd card. With ernd card is the blended price less 10% cash back on Wafer charges. It comes back on the card bill and doesn't change Wafer's list price. It applies when you pay Wafer directly, not through OpenRouter credits.
Questions
What is the cheapest DeepSeek V4.1 Flash 1017 API provider?
Particle AI (US, Flex), at $0.0813 per million tokens blended ($0.025 input, $0.25 output), in OpenRouter's listings on November 7, 2026. If your workload is mostly output, DeepInfra (Flex) has the lowest output price at $0.235.
How much does DeepSeek V4.1 Flash 1017 cost per million tokens?
Input runs from $0.006 to $2.40 per million tokens across 51 listings, and output from $0.15 to $9.60. The middle listing comes to $0.311 blended.
Which DeepSeek V4.1 Flash 1017 provider is fastest?
Baseten (Fast) led on both counts over the last 7 days, at about 545 tokens per second and 0.25 s to the first token.
How do I pay for DeepSeek V4.1 Flash 1017 inference?
Buy OpenRouter credits and it sends each request to one of these providers, or open an account with a provider and pay it directly. The ernd card is a credit card for US companies. It earns 4% back on AI vendor spend and 1.5% back on everything else. Wafer charges on the card earn 10% back.
More models
- DeepSeek V4.1 Flash from $0.0887 blended
- GLM 5.3 Flash from $0.0459 blended
- Hy4 preview from $1.13 blended
- MiniMax M3.1 Flash from $0.131 blended
- DeepSeek V4 Flash 0731 from $0.0667 blended
- GLM 5.3 from $0.402 blended
- Kimi K3.1 Pro from $2.27 blended
- Nemotron 3 Ultra from $0.74 blended
- Hy3 from $0.144 blended
- Nemotron 3.5 Super VL from $0.133 blended
- MiMo-V2.6-Flash from $0.14 blended
- DeepSeek V4 Flash 0423 from $0.0544 blended
- Kimi K3 from $2.61 blended
- Nemotron 3.5 Lightning from $0.0488 blended
- MiMo-V2.6-Pro from $0.486 blended
- GLM 5.2 from $0.29 blended
- DeepSeek V4 Pro 0813 from $0.871 blended
- Qwen3.8 27B from $0.342 blended
- gpt-oss-120b from $0.025 blended
Source
Checked . Providers change prices often, so OpenRouter's page is the final word. ernd is run by Provenance Research Inc.