DeepSeek V4.1 Flash
$0.049 per 1M input tokens and $0.375 per 1M output tokens · 1.05M context · vision, tools, reasoning, JSON, open weights.
Prices checked Oct 21, 2026 · tracked by LLM Cost Index since Oct 21, 2026
Cheaper ways to run the same model: DeepSeek V4.1 Flash (batch) at $0.05 / $0.15 — same model, different service (e.g. batch processing with a delay).
What DeepSeek V4.1 Flash costs per month
Estimated monthly bill for typical workloads, using the cached-input price for repeated context. Try your own numbers in the cost calculator.
| Workload | Per month |
|---|---|
| Customer chatbot 1,000 conversations a day, 8 turns each | $38.77 |
| Document Q&A (RAG) 500 questions a day over retrieved documents | $6.79 |
| Coding / tool-using agent 200 tasks a day, long tool-heavy contexts | $13.68 |
| Content generation 30 long articles a day | $0.976 |
| Classification / extraction 100,000 short items a day | $60.98 |
Price history
We started tracking DeepSeek V4.1 Flash on Oct 21, 2026; the chart appears as soon as there is a second data point. See all recent price changes.
Where to run DeepSeek V4.1 Flash: 38 providers
OpenInference is the cheapest at $0.049 input / $0.375 output per 1M tokens. Uptime is the average over the last 30 days as reported by OpenRouter; see provider reliability.
| Provider | Input / 1M | Output / 1M | Context | Uptime | Notes |
|---|---|---|---|---|---|
| OpenInference | $0.049 | $0.375 | 1.05M | 98.7% | fp4 |
| Relace | $0.07 | $0.50 | 1.05M | 99.8% | — |
| Decart | $0.09 | $0.18 | 1.05M | 100% | fp4 |
| Sail Research | $0.101 | $0.15 | 1.05M | 100% | fp4 |
| Sail Research | $0.101 | $0.15 | 1.05M | 100% | fp4 |
| InferenceNet | $0.103 | $0.715 | 1.04M | 99.5% | fp8 |
| InferenceNet | $0.103 | $0.715 | 1.04M | 99.5% | fp8 |
| DeepInfra | $0.112 | $0.336 | 1.05M | 94.1% | flex tier, fp8 |
| AtlasCloud | $0.135 | $0.54 | 1.05M | 99.9% | fp8 |
| DeepInfra | $0.14 | $0.42 | 1.05M | 100% | fp8 |
| StreamLake | $0.147 | $0.588 | 1.02M | 99.8% | fp8 |
| CoreWeave | $0.15 | $0.65 | 1.05M | 100% | fp8 |
| GMICloud | $0.18 | $0.72 | 1.05M | 99.5% | fp8 |
| Morph | $0.187 | $0.60 | 1.05M | 100% | fp8 |
| Wafer | $0.19 | $0.65 | 1.05M | 99.4% | mxfp4 |
| Novita | $0.195 | $0.78 | 1.05M | 100% | fp8 |
| DeepInfra | $0.21 | $0.63 | 1.05M | 100% | priority tier, fp8 |
| Fireworks | $0.21 | $0.84 | 1.05M | 100% | — |
| Phala | $0.21 | $0.84 | 1.05M | 99.4% | — |
| DigitalOcean | $0.225 | $0.90 | 1.05M | 99.6% | — |
| Makora | $0.26 | $1.17 | 1.05M | 96.9% | fp8 |
| Ionstream | $0.27 | $0.75 | 1.05M | 99.1% | — |
| Crusoe | $0.29 | $1.20 | 1.05M | 100% | fp8 |
| Baidu | $0.30 | $1.20 | 1.05M | 99.9% | fp8 |
| BaseTen | $0.30 | $1.20 | 1.05M | 99.4% | — |
| BaseTen | $0.30 | $1.20 | 1.05M | 99.4% | — |
| Fireworks | $0.30 | $1.20 | 1.05M | 95.2% | — |
| $0.30 | $1.20 | 1.05M | 80.7% | — | |
| Modal | $0.30 | $1.20 | 1.05M | 100% | — |
| Nebius | $0.30 | $1.20 | 1.05M | 99.8% | fp8 |
| Parasail | $0.30 | $1.20 | 1.05M | 98.8% | fp8 |
| PrimeIntellect | $0.30 | $1.20 | 1.05M | 95.3% | — |
| SiliconFlow | $0.30 | $1.20 | 1.05M | 99.9% | fp8 |
| Together | $0.30 | $1.20 | 1.05M | 96.5% | — |
| Venice | $0.30 | $1.20 | 1M | 98.6% | fp8 |
| Alibaba | $0.33 | $1.32 | 1M | 100% | — |
| Fireworks | $0.45 | $1.80 | 1.05M | 100% | — |
| BaseTen | $0.60 | $2.40 | 1.05M | 93.3% | priority tier |
Cheaper alternatives to DeepSeek V4.1 Flash
Models with the same tool calling, image input, structured output for at least 20% less (3:1 input/output mix), most capable first.
| Model | Input / 1M | Output / 1M | Context | Capabilities |
|---|---|---|---|---|
| GLM 5.3 Flash Z.ai | $0.041 | $0.12 | 1.05M | visiontoolsreasoningJSONopen weights |
| Ling 3.0 Flash VL inclusionAI | $0.06 | $0.18 | 131K | visiontoolsreasoningJSONopen weights |
| Qwen3.5-9B Qwen | $0.08 | $0.13 | 262K | visiontoolsreasoningJSONopen weights |
| Ministral 3 3B 2512 Mistral AI | $0.10 | $0.10 | 131K | visiontoolsJSONopen weights |
| Gemma 3 12B | $0.05 | $0.15 | 131K | visiontoolsJSONopen weights |
How to pay less for DeepSeek V4.1 Flash
For a customer chatbot (1,000 conversations a day, 8 turns each) that costs $49.14 a month at list price without caching:
| Option | Per month | You save |
|---|---|---|
| Move to GLM 5.3 Flash (similar capabilities) | $16.82 | $32.32 (66%) |
| Switch to DeepSeek V4.1 Flash (batch) | $30.60 | $18.54 (38%) |
| Use prompt caching (60% of input cached) | $38.77 | $10.37 (21%) |
| Route to Decart | $47.52 | $1.62 (3%) |
Other workloads: cut-your-bill tool for DeepSeek V4.1 Flash.
Specifications
| Model ID | deepseek/deepseek-v4.1-flash |
| Input | text, image |
| Output | text |
| Tool calling | Yes |
| Structured output | Yes |
| Reasoning mode | Yes |
| Open weights | Yes — deepseek-ai/DeepSeek-V4.1-Flash |
| Released | Sep 10, 2026 |
FAQ
- How much does DeepSeek V4.1 Flash cost?
- $0.049 per million input tokens and $0.375 per million output tokens; cached input costs $0.0010 per million. Prices as of Oct 21, 2026.
- What is the context window of DeepSeek V4.1 Flash?
- 1.05M tokens, with up to 944K output tokens per response.
- Does DeepSeek V4.1 Flash support images, tools and structured output?
- Images as input: yes. Tool calling: yes. Structured (JSON) output: yes. Reasoning mode: yes.
- What are cheaper alternatives to DeepSeek V4.1 Flash?
- Models with the same key capabilities for at least 20% less include GLM 5.3 Flash ($0.041 / $0.12), Ling 3.0 Flash VL ($0.06 / $0.18), Qwen3.5-9B ($0.08 / $0.13).
More from DeepSeek: all DeepSeek models · how we collect prices