DeepSeek V4 Flash
DeepSeek-V4-Flash-0731 via Fireworks: official DeepSeek V4 Flash release with stronger agentic performance, served as a serverless MoE at 1M context scale. Function calling supported. Uses 1 base request per send before length multipliers. Does not support web search or image input.
You always get the exact model you pick — we never silently route you to another.
About DeepSeek V4 Flash
DeepSeek V4 Flash bets on a simple proposition: near-Pro coding performance at a fraction of the cost. The official 0731 build is a Mixture-of-Experts architecture with a speculative-decoding module attached, tuned for strong agentic coding and tool use while pricing inputs at $0.14 per million tokens — blending much lower with caching. On coding benchmarks, Flash scores within striking distance of V4 Pro on LiveCodeBench and SWE-bench, making it the preferred choice for most coding tasks in head-to-head comparisons. Users consistently praise the cost-to-performance ratio and the practical 1M token context window for handling full codebases without chunking. The honest caveat: Flash can still stumble on complex multi-step logic — users report syntactically correct code with subtle algorithmic flaws that take several iterations to surface and fix. On just4o.chat it runs via Fireworks serverless as deepseek-v4-flash-0731. For teams running high-volume pipelines or building coding tools where speed and cost matter more than deep reasoning depth, it's a compelling option.
Best for
- High-volume coding assistants and IDE integrations where LiveCodeBench and SWE-bench performance matters at low per-token cost
- Real-time chatbots and customer support systems that need sustained throughput
- Long-document workflows — full codebases, research papers, or technical specs that fit inside the 1M token context window
- Batch automation pipelines (content generation, classification, summarization) where per-request cost is a primary constraint
- Agentic tool-use workflows with native function calling and external API integration
Specifications
| Provider | DeepSeek |
|---|---|
| Released | 2026-07 |
| Intelligence | High |
| Speed | Medium |
| Context window | 1,000,000 tokens |
| Max output | 384,000 tokens |
| Knowledge cutoff | July 2026 |
| Input price | $0.14 / 1M tokens |
| Output price | $0.28 / 1M tokens |
| Request cost | 1 base request |
| Plan tier | Base |
| Input | Text |
| Output | Text |
| Features | Cached input: $0.028 / 1M tokens, 1 base request per send before length multipliers, Function calling supported, Serverless through Fireworks (deepseek-v4-flash-0731) |
| Model ID | deepseek-v4-flash |
Frequently asked questions
On Fireworks serverless, input tokens are $0.14 per million, cached input is $0.028 per million, and output tokens are $0.28 per million.
One million tokens, roughly equivalent to 1,500 pages of standard text.
Flash and Pro sit close on coding benchmarks. Flash is faster and cheaper; Pro is the better choice for tasks requiring deeper or more sustained reasoning.
Real-world testers still catch subtle logical flaws in complex algorithms. It can also generate verbose output, which can inflate costs on high-volume workloads despite the low per-token rate.
DeepSeek V4 Flash maps to accounts/fireworks/models/deepseek-v4-flash-0731, the official July 31 release that replaces the preview serverless endpoint Fireworks is retiring on August 14, 2026.
Yes. V4 Flash is open-weights under an MIT license and available on Hugging Face, with no restrictions on commercial or research use.