Overview Benchmarks Status FAQ Terms Privacy Legal notice Security Support llms.txt 日本語

Benchmarks: GPU seconds by resolution, frames and samples

Measured on JANCTION Render with one NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition (96 GB): a 24-frame 720p MP4 at 64 samples takes about 59 GPU seconds, 48 frames at 1080p/128 samples about 322 s, and a 4-frame preview about 4 s. Nothing is charged during the free beta; the estimated price column uses the planned rate (0.1 JPY per GPU second).

Every number below is copied from the production job log (2026-09-30 to 2026-10-02); the scenes are the simple ones from the guides (a cube, a monkey head, a floor). Heavy scenes (high sample counts, volumes, large textures) scale accordingly. Round trip is submit-to-download with an empty queue.

Cases

kind	input	frames	resolution	samples	lighting	GPU s	round trip	est. price	output	from
preview	bpy script (rotating cube)	4 of 24 (tiled)	640x360 tiles	16	scene lights	4	8 s	0.4 JPY	sheet.png	Claude.ai (remote MCP)
final	bpy script (rotating cube)	24	1280x720	64	scene lights	58.7	65 s	5.9 JPY	output.mp4 (24 fps)	CLI / Claude.ai
final	bpy script (rotating cube)	48	1280x720	64	scene lights	117.3	125 s	11.7 JPY	output.mp4	CLI
final	bpy script (rotating cube)	48	1920x1080	128	scene lights	322.3	335 s	32.2 JPY	output.mp4	Claude Code (stdio MCP)
preview	bpy script (monkey + floor)	1	960x540	16	scene lights	2.4	6 s	0.2 JPY	frame_0001.png	Claude.ai
preview	glTF file (hand-written pyramid) + assets	1	640x360	8	studio (backdrop hidden)	1.8	7 s	0.2 JPY	frame_0001.png	HTTP API
preview	bpy script (torus)	1	640x360	8	night, Blender 5.2	1.9	8 s	0.2 JPY	frame_0001.png	HTTP API
final (EEVEE, webm)	bpy script (spinning monkey + floor)	24	960x540	32	scene lights	25.1	28 s	2.5 JPY	output.webm (VP9)	HTTP API
final (Cycles, mp4)	bpy script (spinning monkey + floor)	24	960x540	32	scene lights	35.9	39 s	3.6 JPY	output.mp4 (H.264)	HTTP API
final (ProRes)	bpy script (spinning monkey + floor)	6	640x360	16	scene lights	5.4	7 s	0.5 JPY	output.mov (ProRes 422 HQ)	HTTP API
preview (EEVEE)	bpy script (spinning monkey + floor)	4 of 24 (tiled)	640x360 tiles	16	scene lights	8.1	10 s	0.8 JPY	sheet.png	HTTP API
preview (compare)	bpy script (monkey + floor)	1 x 4 presets	320x180 tiles	16	studio / sunset / overcast / night	2.6	5 s	0.3 JPY	sheet.png with labels	HTTP API
preview (orbit)	bpy script (monkey + floor)	4 angles (0/90/180/270 deg)	320x180 tiles	8	sunset	2.5	7 s	0.2 JPY	sheet.png	HTTP API
final (orbit)	bpy script (monkey + metal cube)	48 (one full turn)	960x540	24	studio	75.7	80 s	7.6 JPY	output.mp4 (24 fps)	HTTP API

What the outputs look like

A 4-frame preview sheet — 4-frame preview (4 GPU seconds).

The same scene under four lighting presets — One scene under the four environment presets.

48-frame turntable (76 GPU seconds).

Prompts and calls used

preview, 4 of 24 (tiled) @ 640x360 tiles: Using JANCTION Render, build a glossy red cube rotating on a grey floor and show me a 4-frame preview.
final, 24 @ 1280x720: Render 24 frames at 1280x720 as an MP4.
final, 48 @ 1280x720: janction-render render cube_scene.py --frames 1-48 --size 1280x720 --samples 64 --wait
final, 48 @ 1920x1080: Render the final 48 frames at 1080p with 128 samples.
preview, 1 @ 960x540: Show me one preview frame at 960x540.
preview, 1 @ 640x360: POST /v1/files pyramid.gltf, then POST /v1/jobs {kind: preview, environment: studio, environment_visible: false}
preview, 1 @ 640x360: blender="5.2", environment="night"
final (EEVEE, webm), 24 @ 960x540: POST /v1/jobs {kind: final, engine: "eevee", output: "webm", frame_start: 1, frame_end: 24, width: 960, height: 540, samples: 32}
final (Cycles, mp4), 24 @ 960x540: Same scene on Cycles: POST /v1/jobs {kind: final, frame_start: 1, frame_end: 24, width: 960, height: 540, samples: 32}
final (ProRes), 6 @ 640x360: output: "prores"
preview (EEVEE), 4 of 24 (tiled) @ 640x360 tiles: engine: "eevee" (slower than Cycles for a preview: EEVEE compiles shaders for about 7 s in every fresh container; the same 4 frames on Cycles took 2.2 GPU s)
preview (compare), 1 x 4 presets @ 320x180 tiles: POST /v1/jobs {kind: preview, environment: "compare", width: 640, height: 360, samples: 16} (all four presets in one Blender session)
preview (orbit), 4 angles (0/90/180/270 deg) @ 320x180 tiles: orbit=true, environment="sunset"
final (orbit), 48 (one full turn) @ 960x540: render_final(orbit=true, orbit_frames=48, width=960, height=540, samples=24, environment="studio")

How to read the numbers

GPU seconds are the container's run time including Blender start-up and scene loading; the free quota counts exactly this.
render_estimate uses the scene's own earlier measurement when it exists, otherwise the median of recent chunks.
Short jobs are dominated by the fixed start-up (1-2 s); long jobs scale with frames x pixels x samples. EEVEE adds about 7 s of shader compilation per job and then costs about half of Cycles per frame, so it pays off on animations, not on previews.
The price column is a reference computed at the planned rate; as of 2026-10-02 nothing is charged.

Updated 2026-10-02. Operated by JasmyLab Inc..

JANCTION Render v0.4.0 · JasmyLab Inc. · 日本語

Overview Benchmarks Status FAQ Terms Privacy Legal notice Security Support llms.txt 日本語