DeepSeek V4 Flash (Non-reasoning)
AI model by DeepSeek. Real-time pricing and benchmark data.
Benchmarks
Editorial Analysis
Profile. DeepSeek V4 Flash (Non-reasoning) is a budget-tier option from DeepSeek at $0.28/M output tokens. Despite the low price, it scores competitively on GPQA Diamond (71.6%) and IFBench (47.2%) — making it a candidate for high-volume production workloads where unit cost dominates.
Latency. On latency, DeepSeek V4 Flash (Non-reasoning) reports 97 tok/s output speed and 918ms time-to-first-token. Output speed is on the slower end; for real-time chat consider pairing with a streaming front-end or routing latency-sensitive requests to a faster model.
Cost model. For a workload of 1M input tokens + 100K output tokens, DeepSeek V4 Flash (Non-reasoning) costs $0.17 per million input-equivalent requests. Multiply by your monthly request volume to project the bill, and compare against the speed/correctness profile above to decide if the trade-off is worth it for your workload.
Editorial summary generated from public pricing and benchmark data (Artificial Analysis API). All numbers are sourced — see the Artificial Analysis methodology for benchmark definitions. Last refreshed 2026-08-02.
Compare with similar models
| Model | Input | Output | Speed |
|---|---|---|---|
DeepSeek V4 Flash (Non-reasoning)Current | $0.14 | $0.28 | 97 tok/s |
| MiMo-V2.5 | $0.14 | $0.28 | 67 tok/s |
| Gemma 4 12B (Reasoning) | $0.10 | $0.30 | 125 tok/s |
| Gemma 4 12B (Non-reasoning) | $0.10 | $0.30 | 121 tok/s |
| Nemotron 3 Nano Omni 30B A3B Reasoning | $0.07 | $0.30 | 319 tok/s |
| Ling 2.6 Flash | $0.10 | $0.30 | 85 tok/s |