Every line, version by version
Every DeepSeek V4 version in one table: input, output and cached prices, context window, benchmark scores, and what a month of your usage costs on each one.
Prices are read live from the provider pricing pages, so this table cannot go stale.
| Version | Tier | Status | Released | Input | Output | Cached | Context | Blended /1M | Sample /mo |
|---|---|---|---|---|---|---|---|---|---|
4.1 Checkpoint DeepSeek-V4.1-Flash-1017 (released 2026-10-17), served under the API model name `deepseek-flash`; open weights on Hugging Face / ModelScope. The 2026-09-10 launch price is unchanged. The launch checkpoint is kept as its own row (deepseek-flash-0910). Price your own usage | Fast | Active | 2026-10-17 | $0.300 | $1.20 | $0.0060 | 1,000K | $0.750 | Peak hours $0.900Off-peak hours $0.450 |
DeepSeek V4.1-Flash 0910Cheapest 4.1 Launch checkpoint of 2026-09-10 — the one the launch announcement's reference table was measured on. Superseded by the 1017 checkpoint on 2026-10-17; the API no longer serves it (`deepseek-flash` now routes to 1017). Rates are identical to the live checkpoint; kept for long-tail search and for the launch post's score table. Price your own usage | Fast | Legacy | 2026-09-10 | $0.300 | $1.20 | $0.0060 | 1,000K | $0.750 | Peak hours $0.900Off-peak hours $0.450 |
4 Price your own usage | Flagship | Active | — | $1.32 | $3.96 | $0.044 | 1,000K | $2.64 | Peak hours $3.30Off-peak hours $1.65 |
4 | Fast | Legacy | — | — | — | — | — | — | — |
Sample cost = 1M input + 500K output tokens per month at list price. A model with a long-context band is priced on that band — the rule is printed in the row. Use the calculator below for your own workload.
Only the differences we can prove from the pricing and score records — a price that moved, a context window that grew, a benchmark both versions report.
The versions of this line that have a published, traceable score record. 20 benchmarks tracked
DeepSeek V4.1-Flash 1017 leads on 9 of 19 shared benchmarks across 4 models.
| Dimension | DeepSeek V4.1-Flash 1017 | DeepSeek V4.1-Flash 0910 | DeepSeek V4-Pro 0813 | DeepSeek V4-Flash 0731 |
|---|---|---|---|---|
| GPQA Diamond | — | 90.9% | 92.4% | 89.9% |
| HLE | 42.6% | 36.8% | 42.7% | 37.8% |
| Codeforces (Rating) | — | 3471 | 3348 | 3289 |
| Apex Bench v1.1 | 72.1% | — | — | — |
| MathArena Apex | — | 65.6% | 65.3% | 58.6% |
| Terminal-Bench 2.1 | — | 90.6% | 87.9% | 82.7% |
| Terminal-Bench 3.0 | — | 30% | 11.8% | 7.6% |
| Terminal-Bench 4.0 | 42.3% | 31.2% | 12.4% | 7% |
| DeepSWE v1.1 | 77% | 74.2% | 62.7% | 54.4% |
| ProgramBench | 29.1% | 20.3% | 15.5% | — |
| NL2Repo-Bench | 68.2% | 65.4% | 61.5% | 54.2% |
| CyberGym | — | 88.1% | 83.3% | 76.7% |
| SEC-Bench Pro | — | 62.8% | 56.4% | 30.9% |
| ExploitGym | — | 15.3% | 5.4% | 1.8% |
| HLE (w/tools) | — | 63.9% | 60% | 51.5% |
| Automation-Bench | 60.5% | 54.8% | 43.2% | 37.7% |
| Agents' Last Exam | 35.4% | 31.8% | 25.7% | 25.2% |
| Chartography (w/tools) | 83% | 78.9% | — | — |
| BabyVision (w/tools) | 92.5% | 89.6% | — | — |
| ZeroBench-main (w/tools) | 51% | 49% | — | — |
| Blended price | $0.750 /1M | $0.750 /1M | $2.64 /1M | — |
| Full priceInput · Output · Cached input | $0.300 · $1.20 · $0.0060 | $0.300 · $1.20 · $0.0060 | $1.32 · $3.96 · $0.044 | — |
| Context window | 1,000K | 1,000K | 1,000K | — |
Transcribed from the sources named on each leaderboard entry. Coverage differs per version; a missing cell means the source did not report that benchmark.
Adjust the monthly volume — every version of the line recomputes instantly.
List price only: cached input, batch discounts and off-peak rates are not included. The sample column prices 1M input + 500K output tokens per month.