cd /news/artificial-intelligence/show-hn-qwen3-8-27b-api-140-tok-s-on… · home › topics › artificial-intelligence › article
[ARTICLE · art-98830] src=inference.tiyuvta.ai ↗ pub= topic=artificial-intelligence verified=true sentiment=· neutral

Show HN: Qwen3.8-27B API, 140 tok/s on one GPU

Tiyuvta AI launched a hosted API for the Qwen3.8-27B model, priced at $0.38 per million input tokens, $0.20 per million cached tokens, and $2.60 per million output tokens, with a throughput of 140 tokens per second on a single GPU. New accounts receive $10.00 in free credit, with 93 of 100 places still available.

read1 min views1 publishedAug 16, 2026

account

One step. The same screen creates the account, so there is nothing to confirm afterwards and no waiting list.

free credit$10.00 of credit is added to your account on sign-up. 93 of 100 places left.

  • Base URL https://api.tiyuvta.ai/v1

  • Models qwen/qwen3.8-27b

· Step-3.7-Flash in bring-up- Price

  • $0.38 input · $0.20 cached · $2.60 output per million tokens
  • Billing
  • Prepaid credit. At zero balance the API returns 402

.

or

By continuing you accept the [terms](/legal/terms) and the
[privacy terms](/legal/privacy).
source & further reading
inference.tiyuvta.ai — original article
── more in #artificial-intelligence 4 stories · sorted by recency
── more on @tiyuvta ai 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
→ Live at https://your-agent.zahid.host ✓
Get free account → Pricing
from €0/mo · no card required
LIVE [news/show-hn-qwen3-8-27b-…] indexed:0 read:1min 2026-08-16 · —