DeepSeek V4.1 Flash 1017
Summary
DeepSeek V4.1 Flash 1017 is an open-weight language model from DeepSeek, released on 17 Oct 2026. It accepts text and images and generates text, with a 1M-token context window and up to 393K output tokens per response. As of 18 Oct 2026, the lowest listed API price is $0.07 per million input tokens and $0.30 per million output tokens (Vercel AI Gateway), across 5 providers we track.
October 17 update of DeepSeek V4.1 Flash with improved reasoning, coding, agentic tasks, and visual understanding
Last verified 18 Oct 2026. Every figure on this page links to its source.
Specifications
| Developer | DeepSeek |
|---|---|
| Release date | 17 Oct 2026 [models.dev] |
| Status | Generally available [models.dev] |
| Weights | Open weights [Hugging Face] |
| License | mit [Hugging Face] |
| Parameters | 763B [Hugging Face] |
| Context window | 1M (1,000,000 tokens) [models.dev] |
| Max output | 393K tokens [models.dev] |
| Knowledge cutoff | May 2025 [models.dev] |
| Input | Text and images [models.dev] |
| Output | Text [models.dev] |
| Reasoning mode | Yes [models.dev] |
| Tool calling | Yes [models.dev] |
| Structured output | Yes [models.dev] |
| Hugging Face | deepseek-ai/DeepSeek-V4.1-Flash-1017 [Hugging Face] |
DeepSeek V4.1 Flash 1017 API pricing by provider
USD per million tokens. Sorted by input price.
| Provider | Input | Output | Cached input | Context | Source |
|---|---|---|---|---|---|
| Vercel AI Gateway deepseek/deepseek-v4.1-flash-1017 | $0.07 | $0.30 | $0.0015 | 1M | models.dev, 18 Oct 2026 |
| DeepSeek API deepseek-flash | $0.15 | $0.60 | $0.003 | 1M | models.dev, 18 Oct 2026 |
| OpenRouter deepseek/deepseek-v4.1-flash-1017 | $0.15 | $0.60 | $0.003 | 1M | OpenRouter, 18 Oct 2026 |
| DeepInfra deepseek-ai/DeepSeek-V4.1-Flash-1017 | $0.20 | $0.60 | $0.006 | 1M | models.dev, 18 Oct 2026 |
| Baseten deepseek-ai/DeepSeek-V4.1-Flash-1017 | $0.30 | $1.20 | – | 1M | models.dev, 18 Oct 2026 |
| Vercel AI Gateway deepseek/deepseek-v4.1-flash-1017-fast | $0.60 | $2.40 | $0.01 | 1M | models.dev, 18 Oct 2026 |
Running DeepSeek V4.1 Flash 1017 locally
Estimated memory, from the parameter count and the architecture in config.json on Hugging Face. Includes about 0.8 GB of runtime overhead.
| Quantization | Weights | Total, 8K context | Total, 32K context |
|---|---|---|---|
| Q4_K_M (4-bit) | 463 GB | 464 GB | 466 GB |
| Q8_0 (8-bit) | 811 GB | 812 GB | 814 GB |
| FP16 / BF16 (16-bit) | 1526 GB | 1528 GB | 1530 GB |
At 4-bit with an 8K conversation, DeepSeek V4.1 Flash 1017 needs about 464 GB, more than any single consumer graphics card; it needs a unified-memory machine, a multi-GPU setup or a split between graphics card and system RAM. Check your own hardware, or see Best computer for local AI on our sister site AIPCs.
Frequently asked questions
What is the context window of DeepSeek V4.1 Flash 1017?
DeepSeek V4.1 Flash 1017 has a context window of 1,000,000 tokens (1M) and can generate up to 393,216 tokens in a single response, according to models.dev.
How much does the DeepSeek V4.1 Flash 1017 API cost?
Through DeepSeek API, DeepSeek V4.1 Flash 1017 costs $0.15 per million input tokens and $0.60 per million output tokens. The lowest listed price is $0.07 input / $0.30 output through Vercel AI Gateway. Prices last verified 18 Oct 2026.
Is DeepSeek V4.1 Flash 1017 open source?
DeepSeek V4.1 Flash 1017 is an open-weight model: its weights are published on Hugging Face as deepseek-ai/DeepSeek-V4.1-Flash-1017 under the mit license. Check the license terms before commercial use.
When was DeepSeek V4.1 Flash 1017 released?
DeepSeek released DeepSeek V4.1 Flash 1017 on 17 Oct 2026, according to models.dev.
Which providers offer DeepSeek V4.1 Flash 1017?
We track DeepSeek V4.1 Flash 1017 on 5 providers: Vercel AI Gateway, DeepSeek API, OpenRouter, DeepInfra, Baseten.
How much VRAM does DeepSeek V4.1 Flash 1017 need?
About 464 GB at 4-bit (Q4_K_M) with an 8K-token conversation, and about 812 GB at 8-bit. That is more than any single consumer graphics card holds. These are estimates from the model's published parameter count and architecture.
Change history
- License changed from none to mit
- Parameter count changed from none to 763B
- Now listed on OpenRouter
- Now listed on Baseten
- Added to the directory
- Now listed on Vercel AI Gateway
- Now listed on Vercel AI Gateway
- Now listed on DeepInfra
- Now listed on DeepSeek API
Sources
- https://openrouter.ai/deepseek/deepseek-v4.1-flash-1017
- https://github.com/sst/models.dev/blob/dev/providers/deepseek/models/deepseek-flash.toml
- https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017
- https://github.com/sst/models.dev/blob/dev/providers/vercel/models/deepseek/deepseek-v4.1-flash-1017.toml
- https://github.com/sst/models.dev/blob/dev/providers/deepinfra/models/deepseek-ai/DeepSeek-V4.1-Flash-1017.toml
- https://github.com/sst/models.dev/blob/dev/providers/baseten/models/deepseek-ai/DeepSeek-V4.1-Flash-1017.toml
- https://github.com/sst/models.dev/blob/dev/providers/vercel/models/deepseek/deepseek-v4.1-flash-1017-fast.toml