ChainThink report, September 11: According to Artificial Analysis testing, DeepSeek V4.1 Flash's comprehensive Intelligence Index score is 40, surpassing its own V4 Pro 0813 at 36, but still below Kimi K3's 44 and GLM-5.3's 45.

According to this third-party ranking, DeepSeek has not yet reclaimed the top spot among domestic models. The cost and speed gap remains significant.

V4.1 Flash averages $0.27 per Intelligence Index task, while Kimi K3 and GLM-5.3 both hover around $2, resulting in a ~7x difference;

Output speed reaches 197 Token/s, far exceeding the ~36 Token/s and ~58 Token/s of the two competitors. Agent performance remains a key strength.

On AutomationBench-AA, V4.1 Flash achieved 69%, tying with GPT-6 Astra and surpassing GLM-5.3’s 62%; on long-context evaluation AA-LCR, it scored 84%.

However, its output length is relatively high—average output per task on AA tests is approximately 89,000 Tokens, 62% more than V4 Pro, partially offsetting its pricing advantage.