DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, with 13B active parameters out of 284B total. This re-post-trained revision is suited for coding, reasoning, and agent workflows....
At a glance
- Context window
- 1.3Mtokens
- Max output
- 944Ktokens
- Knowledge to
- —
- Released
- 2026
Capabilities
Benchmarks
- Intelligence index
- 35
- Coding index
- 69
- Agentic index
- 42
Design Arena
| Category | Elo | Win rate |
|---|---|---|
| models · uicomponent | 1,252 | 47 |
| models · website | 1,251 | 46.8 |
| models · codecategories | 1,246 | 46.5 |
| models · 3d | 1,237 | 48.6 |
| models · gamedev | 1,233 | 44.7 |
| models · svg | 1,212 | 45.1 |
| models · dataviz | 1,197 | 41.2 |
| models · asciiart | 1,110 | 33.3 |
Standard evaluations
| Benchmark | Score | Tasks |
|---|---|---|
| gpqa_diamond | 0.85 | 1,188 |
| tau_bench_verified_airline | 0.73 | 300 |
Source: Artificial Analysis, Design Arena · as of Sep 16, 2026
Performance
Uptime on IranRouter
8.1%
Reference uptime
100%
Median latency
1,203 ms
Throughput
51.3 tok/s
IranRouter uptime comes from our own automated probes; the reference figures are daily measurements of this model's upstream routes. Each card shows a 30-day average, and a day without data stays empty.
Popularity
- Current rank
- #4
- Rank change, 7 days
- 0
Daily tokens · 90 days
1.5T
Ranked by daily token volume across the global model market; the figure above the chart is the daily average.
Supported parameters
Parameters this model accepts; anything else is ignored.
Quick start
from openai import OpenAI
client = OpenAI(
api_key="ir-...",
base_url="https://iranrouter.com/v1",
)
resp = client.chat.completions.create(
model="deepseek/deepseek-v4-flash-0731",
messages=[{"role": "user", "content": "سلام!"}],
)
print(resp.choices[0].message.content)Only base_url and the key change; the rest of your code stays as it is.