Kling 3.0 Turbo
Video Available Full comparison ↗ klingai/kling-3-0-turbo · by Kuaishou Technology
· Diffusion transformer (DiT)
Pricing — 1 offering(s)
720p (native audio included)
- $0.112 / second
Current
2026-09-20 → present
0.8 Units/s at $0.14/unit list price. Matches the exact figure Kling's own changelog gave at Turbo's…
0.8 Units/s at $0.14/unit list price. Matches the exact figure Kling's own changelog gave at Turbo's 2026-06-17 launch ("0.8 units are deducted per second for 720P generation") — unchanged since launch.
1080p (native audio included)
- $0.14 / second
Current
2026-09-20 → present
1.0 Unit/s. Matches the launch-day changelog figure ("1.0 unit is deducted per second for 1080P generation")…
1.0 Unit/s. Matches the launch-day changelog figure ("1.0 unit is deducted per second for 1080P generation") — unchanged since launch.
Showing the active price and any recorded history. Full pricing history is available via the paid API — see API docs.
See how Kling 3.0 Turbo fits into a cost-aware routing setup
See how →Capability profile
How Kling 3.0 Turbo rates across core capability dimensions, with the task-level evidence behind each rating.
Positioned by Kling as the value/speed tier; not claimed to match base Kling 3.0's quality ceiling.
Native audio is always included at a single flat rate — no audio-off tier exists, unlike base Kling 3.0.
3-15 seconds per call.
Max 1080p via API — no 4K tier, unlike Kling 3.0 and Kling 3.0 Omni.
Explicitly marketed as the faster-output variant of the Kling 3.0 line.
$0.112-0.14/s (720p/1080p, audio included) — cheapest of the three Kling 3.0 models at matching resolutions.
No reference-video input (text/image input only, per Kling's capability map) — no video-to-video editing capability.
Ratings are estimated — limited independent data is available for this model.
Operator guidance
Choose Kling 3.0 Turbo over base Kling 3.0 for lower cost and faster output when 4K resolution and reference-video input aren't needed. Step up to Kling 3.0 for 4K, or Kling 3.0 Omni for video-input-driven editing.
Use cases
- Cost- and latency-sensitive text/image-to-video generation
- High-volume production workflows where 4K isn't required
Provenance notice
- Chinese-origin model; subject to Chinese data regulations and content moderation policies
- Content policy restrictions may differ from Western providers for certain topics
Limitations
- No 4K resolution support
- No reference-video / video-to-video input
- No published third-party benchmark scores as of 2026-09-20