Qwen3.8 27B
Dense 27B vision-language model for coding, agent tasks, and image and video understanding
Context
1,131,072
Max output
262,144
Input / 1M
$0.10
Output / 1M
$0.40
Providers
28
01Profile
Profile
Released
2026-08-14
Last updated
2026-09-02
Knowledge cutoff
—
Weights
Open
02Capabilities
Capabilities
Input
text · image · video · pdf
Output
text
Reasoning
Yes
Tool calling
Yes
Structured output
Yes
Attachments
Yes
Cache read
—
Cache write
—
03Pricing
Pricing across 29 providers
Per 1M tokens, USD. Sorted by input price. First-party row highlighted.
04History
Change history
12+ recent events since 2026-08-14
2026-09-15Price cutdeepinfrainput / 1M$0.40→ $0.20
2026-09-15Price cutdeepinfraoutput / 1M$3.00→ $2.50
2026-09-15Price cutovhcloudinput / 1M$0.46→ Not listed
2026-09-15Price cutovhcloudoutput / 1M$3.11→ Not listed
2026-09-14Addedamdoffering→ new
2026-09-14Price upovhcloudinput / 1MNot listed→ $0.46
2026-09-14Price upovhcloudoutput / 1MNot listed→ $3.11
2026-09-12ContextkiloContext window1,000,000→ new
2026-09-12Price cutopenrouterinput / 1M$0.42→ $0.21
2026-09-12Price cutopenrouteroutput / 1M$3.00→ $2.55
2026-09-11Price cutcortecsinput / 1M$0.33→ $0.10
2026-09-11Price cutcortecsoutput / 1M$2.45→ $0.40