by Alibaba (Qwen Team)

Qwen3 235B A22B Instruct 2507.

text open weights datacenter 235B params 262K ctx
Cheapest input
$0.09/M
on DeepInfra
Cheapest output
$0.55/M
on DeepInfra
Smallest GPU
4× Nvidia Q RTX 8000
$1.08/hr
API pricing

What it costs per month across providers.

Estimate your monthly bill for Qwen3 235B A22B Instruct 2507 across every host that publishes per-token pricing. Slide your token volumes; the chart + table re-rank cheapest-first.

Cheapest
$2.0
DeepInfra
Most expensive
$3.0
OpenRouter
Spread
$1.0
max − min
Providers
2
with priced rows
100,000 1,000,000,000
100,000 500,000,000
Monthly bill

Cheapest provider on the left.

Total monthly cost — input + output tokens combined.

Loading...
Per provider

Bill breakdown.

Provider Monthly total
$2.0 Sign up ↗
$3.0 Sign up ↗
Self-host alternative

Rent the GPU instead of paying per token.

Full breakdown →

For an open-weights model like Qwen3 235B A22B Instruct 2507, you can rent a GPU and serve inference yourself. The math: cheapest GPU rental × 730 hours/month. Power is included in a rental's hourly price.

GPU rental
$789.57
4× Nvidia Q RTX 8000 @ $1.08/hr
Self-host total
$789.57
per month
API beats self-host
+$787.57
API is cheaper here

Assumes the GPU runs 24/7 at ~85% utilisation. If your traffic is bursty, you'll pay less for the API and probably more for the GPU (idle hours still cost rental). The breakeven analysis lives on the Self-host vs API breakeven tool.

Full calculator
Want to compare token volumes across other models too? Open the standalone API pricing tool →