DeepSeek-V4 prompt KV cache is lost after short idle despite identical prefix

Important limitations

• This may be one public report, not widespread demand.
• Current independent report count: 1.
• This is currently a single independent report. It has not passed clustering or human review as a verified signal.
• No market size, willingness to pay, competition, or trend claims are inferred here.

What was reported

Who is affected

Application developer using the reported software

Job to be done

Configure an application runtime reliably

Workflow

I tested the bundled runner with the following environment variables set before starting Ollama:

Desired outcome

Since the model remains loaded and the request prefix is identical, I expected the prompt KV cache/checkpoint to remain reusable after the idle period. Only the newly appended input should require processing.

Severity

Not enough evidence yet.

Urgency

high

Evidence indicators

What would upgrade this

• Additional independent reports from distinct communities or authors.
• Human clustering into a review signal, then explicit publication approval.
• Counter-evidence review and durable original-source verification.

Original source

Open the original public report