Deepseekv4.1 Flash
DeepSeek V4.1 Flash 1017 is the October 17, 2026 snapshot of DeepSeek V4.1 Flash, served by DeepSeek under the `deepseek-flash` API model name.
deepseekv4.1-flash-1017Released 17 Oct 2026 (5 weeks ago)Pricing
List prices at Sarvam, in US dollars per million tokens.
| Tier | Input / 1M | Output / 1M | |
|---|---|---|---|
| Standard | $0.149 | $0.585 | |
| Cached input 98% off input | $0.00266 | 98% off input |
What it costs a month
At Sarvam list prices, 30 days.
- Chat assistant5,000 requests a day, 2K in / 0.5K out, cached$75.40
- RAG over documents2,000 requests a day, 8K in / 0.6K out, cached$57.45
- Coding agent1,000 requests a day, 40K in / 2K out, cached$90.96
- Overnight batch20,000 requests a day, 4K in / 0.4K out$498
Where to run it
1 host serve Deepseekv4.1 Flash. The comparison is on blended price, 3 parts input to 1 part output.
| Host | In / out per 1M | Model id | |
|---|---|---|---|
| $0.149 / $0.585 | deepseekv4.1-flash-1017 |
Capabilities
Core
- Streaming
- Function calling
- JSON mode
- Structured outputs
Multimodal
- Vision
- Image input
- Audio
- Image generation
- Video
Advanced
- Reasoning
- Tool calling
- MCP
- Prompt caching
- Batch API
- Embeddings
- Takes in
- TextImages
- Produces
- Text
- Features
- Parallel tool callsPrompt caching
- Open weights
- License: DeepSeek License (MIT code)Runs in about 1831.7 GB of GPU memoryRuntimes: SGLang, TGI, vLLM
Cheaper alternatives
Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.
Same model, cheaper host
No host sells it for less than Sarvam.
Smaller DeepSeek models
No cheaper DeepSeek model keeps its key capabilities.
Other frontier labs
- 22% less
- 22% less
Open-weight picks
- 2% less
- 31% less
- 45% less
- 46% less
Call it
Using Sarvam's API.
from litellm import completion
response = completion(
model="sarvam/deepseekv4.1-flash-1017",
messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)Quick answers
How much does Deepseekv4.1 Flash cost?
On Sarvam, Deepseekv4.1 Flash costs $0.149 per million input tokens and $0.585 per million output tokens. Cached input costs $0.00266 per million tokens.
Which APIs offer Deepseekv4.1 Flash?
Deepseekv4.1 Flash runs on 1 API: Sarvam.
What is Deepseekv4.1 Flash's context window?
1M tokens, with up to 1M tokens in one reply.
Is Deepseekv4.1 Flash being retired?
No retirement date is scheduled for it on any API we track.
What are cheaper alternatives to Deepseekv4.1 Flash?
Models that cost less and keep most of what it can do: Claude Haiku 5.5 on Anthropic (22% less), GPT-6 Luna on OpenAI (22% less) and Qwen3.8 27B on EmpirioLabs AI (2% less). Savings are on a mix of 3 input tokens to 1 output token.
When was Deepseekv4.1 Flash released?
DeepSeek released it on 17 Oct 2026.
What is Deepseekv4.1 Flash's API model id?
Use deepseekv4.1-flash-1017 with Sarvam.
Sources
Raw catalog entry for deepseekv4.1-flash-1017 on Sarvam
{
"cache_read_input_token_cost": 2.659574468e-9,
"comment": "These are Sarvam's introductory rates, billed through the end of 2026. Update on 2026-12-31 when the promotion ends. The last published list rates per 1M tokens are INR 25 input, INR 0.5 cached input and INR 105 output (https://docs.sarvam.ai/changelog/2026/11/7.md).",
"input_cost_per_token": 1.489361702e-7,
"litellm_provider": "sarvam",
"max_input_tokens": 1048576,
"max_output_tokens": 1048576,
"mode": "chat",
"output_cost_per_token": 5.85106383e-7,
"source": "https://www.sarvam.ai/token-factory/pricing",
"supports_audio_input": false,
"supports_function_calling": true,
"supports_parallel_function_calling": true,
"supports_prompt_caching": true,
"supports_reasoning": true,
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_video_input": false,
"supports_vision": true
}