DeepSeek V4 Flash Vision Exp Official Specifications, Pricing & Benchmarks
DeepSeek V4 Flash Vision Exp (deepseek-v4-flash-vision-exp) by DeepSeek, released 2026-10-17, featuring 1,048,576 context window, 393,216 max output tokens, and Text+Image input.
📋 DeepSeek V4 Flash Vision Exp Official API & System Card Specifications
Verified against official DeepSeek developer documentation & pricing
1. Identity & Capabilities
API Spec- API Model ID
- deepseek-v4-flash-vision-exp
- Release Date
- 2026-10-17
- Weight License
- Open-Weight (API Spec)
- Native Reasoning
- ✅ Supported
- Function Calling & JSON Schema
- ✅ 原生支持
- Batch API 异步批处理
- ✅ Supported (50% off)
2. Context, Output & Modalities
Limits- Max Context Window
- 1,049,000 tokens (1049K)
- Max Output Tokens
- 393,216 tokens
- Input Modalities
- Text + Image
- Output Modalities
- Text
- Direct China Access
- ✅ Direct CN + RMB
- Deployment Mode
- Open-Weight + Cloud API
3. Official Pricing & Cache
USD / 1M- Input / 1M Tokens
- $ 0.15
- Cached Input / 1M
- $ 0.003 -98%
- Output / 1M Tokens
- $ 0.6
🎛️ Official Sampling & Reasoning Parameters
Recommended generation settings & API parameter constraints
📊 DeepSeek V4 Flash Vision Exp Full Benchmark Scorecard
7 public benchmarks · ranks among 154 models · synced 2026-11-20
💻 Coding 4 items
Terminal-Bench 2.1 is an updated release of the Terminal-Bench benchmark that tests AI agents' ability to operate a computer via the terminal.
DSBench-Hard is DeepSeek's internal test set of difficult coding-agent problems.
DeepSWE is a software engineering agent benchmark evaluated with the mini-swe-agent harness, where each task is solved in an isolated container with no internet access. It measures an agent's ability to autonomously resolve real-world coding issues end to end.
NL2Repo evaluates long-horizon coding capabilities including repository-level understanding, where models must generate or modify code across entire repositories from natural language specifications.
🤖 Agents & Tools 2 items
Agents' Last Exam is a challenging benchmark for AI agents on hard, long-horizon tasks that test sustained reasoning, planning, and tool use, reported with and without tool access.
AutomationBench is a tool-use benchmark that evaluates AI agents on automating real-world workflows, testing their ability to orchestrate tools and complete multi-step automation tasks.
Source: ZeroEval / llm-stats.com (✓ = third-party verified, others vendor-reported).
Closest Self-Hostable Open-Weight Alternatives to DeepSeek V4 Flash Vision Exp
Open-weight models with full HuggingFace weights for private GPU / Mac deployment
Frequently Asked Questions
Is DeepSeek V4 Flash Vision Exp open-source or proprietary?
DeepSeek V4 Flash Vision Exp (deepseek-v4-flash-vision-exp) is a proprietary commercial model by DeepSeek. Weights and internal layer dimensions are undisclosed; it is accessed via cloud API.
What are the context limit and API pricing of DeepSeek V4 Flash Vision Exp?
DeepSeek V4 Flash Vision Exp supports 1049K input context and 393,216 max output tokens, priced at $0.15/1M input ($0.003/1M cached) and $0.6/1M output.