Qwen3.8 Flash API: pricing, base URL and quick setup

Specifications and pricing

Prices synced with platform billing (USD per 1M tokens).

Quick setup

curl https://a2agent.me/v1/chat/completions \
  -H "content-type: application/json" \
  -H "authorization: Bearer sk-your-a2agent-key" \
  -d '{
    "model": "qwen3.8-flash",
    "messages": [{"role": "user", "content": "Reply with OK"}]
  }'

Using it from a coding client

A good first-pass model in a two-tier setup: Flash reads and filters, a MAX model handles what survives. Both take the same key and the same base URL, so the switch is one string.

• Run it in Cline
• Run it in Kilo Code

How it compares within its family

Qwen3.8 Flash lists at $0.15 per million input tokens and $0.47 output, the lowest of the three Flash releases here. Qwen3.7 Flash is $0.20 and $0.80; Qwen3.6 Flash is $0.20 and $1.20. All three carry the vision tag and all three run the 1M window shared across Qwen, so the newer release is cheaper without giving up context or modality, which makes the older two hard to justify for new work. Against the MAX tier the gap is more than tenfold on input. What Flash does not carry is the reasoning tag, and for long chains of inference the MAX models are the ones tuned for it. For reading, extraction and triage at volume, this is the rate to start from.

Other models on this platform

• Qwen3.5 Plus1M
• Qwen3.6 Flash1M
• Qwen3.7 Flash1M
• Qwen3.7 Plus1M
• Qwen3.7 MAX1M

Questions