DeepSeek
DeepSeek V4 Flash
DeepSeek V4 Flash is a DeepSeek model with 1M context, Not verified in / Not verified out per million tokens, last verified October 1, 2026. Open weights: yes. Status: Retired since September 10, 2026; the current pick is DeepSeek V4.1 Flash. Public-board scores are linked to the publisher; vendor-only figures stay in the watch-out, not on the board.
DeepSeek V4 Flash was the efficient open-weight (MIT) sibling of V4 Pro — retired from DeepSeek's API on 2026-09-10 and replaced by V4.1-Flash.
Recorded profile verification October 1, 2026Individual provider fields may change
| Specification | DeepSeek V4 Flash |
|---|---|
| Provider | DeepSeek |
| Tier | Balanced |
| Context window | 1M |
| Max output | Not verifiedUnverified |
| Input / 1M tokens | Not verifiedUnverified |
| Output / 1M tokens | Not verifiedUnverified |
| Weights | Open |
| Parameters | 284B total / 13B active (MoE) |
| Reasoning levels | low, high, max |
| Modalities | text |
| License | MIT |
| API model id | deepseek-v4-flash |
| Released | April 24, 2026 |
Pricing tiers: Retired 2026-09-10 — no longer sold. Historical checkpoint tariffs are not verified here. The legacy deepseek-v4-flash ID temporarily routes to V4.1-Flash at $0.30/$1.20 peak or $0.15/$0.60 off-peak. MIT weights remain. A current 384K API output limit belongs to the routed V4.1-Flash; the original checkpoint's hard output cap is not verified.
Publisher evidence
Published benchmark results
Each result keeps its source date visible. Dates refer to source publication or capture unless the source explicitly states when it measured the result. A missing benchmark is not treated as a zero.
| Benchmark | Score | Source date | Source |
|---|---|---|---|
| SWE-bench Verified | 79% | 2026-04-24 | DeepSeek original family HF model card (vendor, Think Max; historical April checkpoint, not the 0731 refresh) · View source |
| GPQA Diamond | 88.1% | 2026-04-24 | DeepSeek family HF model card (vendor, Think Max, pass@1) · View source |
| Humanity's Last Exam | 45.1% | 2026-04-24 | DeepSeek family HF model card (vendor, Think Max, with tools; 34.8 no tools) · View source |
| MMLU-Pro | 86.2% | 2026-04-24 | DeepSeek family HF model card (vendor, Think Max, exact match) · View source |
Best for
- Open-weight deployments
- Cost-sensitive inference
- High-volume serving
Watch out
History
DeepSeek V4 Flash timeline
Release and status events for this model that passed two-source verification — dated, sourced, and linked.
- DeepSeek V4-Flash retiredStatus updateSeptember 10, 2026
DeepSeek retires V4-Flash and V4-Flash-Vision-Exp. The legacy ids temporarily route to V4.1-Flash and bill at V4.1-Flash rates.
DeepSeek — V4.1-Flash release notes - DeepSeek V4-Flash-Vision-ExpLimited previewAugust 21, 2026
Experimental image-input variant of V4-Flash on the DeepSeek API.
DeepSeek API changelog - DeepSeek-V4-Flash-0731ReleaseJuly 31, 2026DeepSeek API changelog
Source receipts
Catalog figures for this model were checked against the following sources.
- Whole-profile primary source check October 1 (accessed 2026-10-01)
- Whole-profile primary source check October 1 (accessed 2026-10-01)
- DeepSeek — V4 news (accessed 2026-08-29)
- DeepSeek API pricing (accessed 2026-08-29)
- DeepSeek — V4.1-Flash release (V4-Flash retired) (accessed 2026-09-26)
- DeepSeek API pricing (legacy id billed at V4.1-Flash rates) (accessed 2026-09-26)
Common questions
DeepSeek V4 Flash
Answered from the verified figures on this page rather than general guidance.