DeepSeekproprietary
Released: 2026-08-01DeepSeek V4.1 Flash
Foundation model evaluated on LiveBench September 2026 by DeepSeek.
Intelligence Index
93.5/ 100
Calibrated multi-domain compositeThroughput Speed
96tok/sec
Output streaming throughputTime To First Token
290ms
Initial chunk server latencyLiveBench Contam-Free
83.9%
September 2026 suite scoreLiveBench Multi-Domain Evaluation Breakdown
Monthly refreshed contamination-free test setsLogical Reasoning88.4%
Coding & Repo Repair85.1%
Mathematics (AIME)89.0%
Data Analysis & Tables80.5%
Language & Comprehension78.5%
Instruction Following82.0%
Sourced Benchmark Evaluations (3)
Verified performance across authoritative benchmarks with provenance tracking.OpenAI Compatible API ExecutionAPI Ready
from openai import OpenAI
client = OpenAI()
response = client.chat.completions.create(
model="deepseek-v4-1-flash",
messages=[{"role": "user", "content": "Evaluate multi-step logic problem"}],
temperature=0.2
)
print(response.choices[0].message.content)Evaluation & Sourcing Notes
Scores listed for DeepSeek V4.1 Flash represent verified evaluations extracted from official research papers, independent evaluation suites (HELM, LMSYS, OpenCompass, LiveBench), and verified audit reports.
All benchmarks marked as saturated or deprecated reflect historical performance where the benchmark no longer provides active discriminative power.