DeepSeek
DeepSeek-V4.1-Flash-1017
Launch posts: Introducing HIPAA-compliant DeepSeek-V4.1-Flash-1017 · Introducing HIPAA-compliant DeepSeek-V4.1-Flash
Specs
- Input price
- $0.30/M
- Cached input price
- $0.006/M
- Output price
- $1.20/M
- Context window
- 1M tokens
DeepSeek's October 17 checkpoint for the deepseek-flash endpoint - the same 552B multimodal MoE, now with Terminal-Bench 4.0 at 42.3, DeepSWE v1.1 at 77.0, and a three-position reasoning_effort switch.
Capabilities
- 552B MoE (1 shared + 384 routed experts, 6 active per token) split into a 20-layer causal encoder and 20-layer decoder
- Terminal-Bench 4.0 at 42.3 and DeepSWE v1.1 at 77.0 - 11.1 and 2.8 points above the September checkpoint
- Automation-Bench 60.5, Agents' Last Exam 35.4, and Apex Bench 72.1 lead DeepSeek's comparison table
- Native vision in the same checkpoint: Chartography with tools 83.0, BabyVision 92.5, ZeroBench-main 51.0
- 1M-token context, 384K max output, and reasoning_effort as a low/high/max switch
Best for
- Input-heavy clinical document pipelines: prior-auth packets, chart exports, and claims history
- Long-horizon automation agents: claim assembly, referral routing, and back-office workflows
- Agentic coding and security-adjacent review at Flash-class cost
Limitations to keep in mind
- Hard reasoning still trails the frontier: HLE 42.6 vs Opus 5.5's 64.4 and GPT-5.6 Sol's 44.5
- Terminal-Bench 4.0 at 42.3 trails Opus 5.5's 66.4 and Opus 5's 51.8
- Verbose at max effort: 210M generated tokens on Artificial Analysis' index, above the 140M median
- The effort dial changed shape - integer reasoning_effort values no longer mean what they did on the previous checkpoint
- Vision reads documents, not diagnostics; no radiology or pathology claim
HIPAA-compliant hosting
DeepSeek-V4.1-Flash-1017 is available under our HIPAA-compliant design-partner program, with a signed Business Associate Agreement (BAA), encryption in transit and at rest, and access controls. We are currently onboarding design partners, with general availability coming soon. Run the model behind a unified OpenRouter-style API and swap to another model with a single parameter. For the deployment requirements behind that, see our 2026 guide to HIPAA compliance for AI & LLM APIs.
Pricing in context
Open source models like DeepSeek-V4.1-Flash-1017 can be served efficiently on optimized inference infrastructure, with savings passed through to you. Exact savings depend on the model and your volume, but open source inference is typically a fraction of the per-token cost of closed-source frontier models on Azure OpenAI or AWS Bedrock - without cloud egress lock-in or minimum commitments. See the full pricing comparison or browse the model catalog.
Design partner program
Run DeepSeek-V4.1-Flash-1017 under HIPAA
Join the waitlist to be prioritized. We'll reach out with a qualification call and early access.