Bulk Queue

Queue translation runs with two-phase uploads and resilient processing.

No Live RunSelect a run to watch live updates.

Queue Health

Loading queue health…

Submit New Run

Source LanguageTarget LanguageConcurrent filesThis backend currently allows a single file to run at once. Current setting: 5 files. OpenRouter requests are also governed by backend per-file batch and rate-limit controls, so higher concurrency increases throughput and API pressure without bypassing provider safeguards.

Contemporary English rewrites archaic or formal English into clear present-day English.

Narrative body prose should change meaningfully. Short labels, proper nouns, copyright/front matter, and lines that already read naturally in modern English may remain close to the source.

Production-ready

Deep review, publishable audit, and canonical promotion are fully hardened on this path.

Canonical publishable destination: Publishable Outputs/Contemporary English

Queue submission policy

Waiting for live queue submission policy from the backend before allowing new runs.

The start controls stay disabled until the backend clears the active blocker.

Translation Model

Selected model

Gemma 4 31B (OpenRouter) — default

Provider / host

OpenRouter

Model ID

google/gemma-4-31b-it

Default

Yes

Cost

Paid (~$0.13/M in · $0.38/M out)

OpenRouter API key override (optional)OpenRouter Gemma is the only supported production runtime. The server key is used by default unless you paste a different OpenRouter key here.

OpenRouter preset selection is disabled. Use a manual key or server `OPENROUTER_API_KEY`.

Dry run (no model calls, cost-free pipeline simulation)Reuse cached translations from prior runsDisabled by default for fresh runs so progress reflects new model work. Resume and Retry Failed reuse prior work automatically.

New upload runs always start fresh. Identical files are no longer redirected into an older run.

Choose Files

Select one or more .docx manuscripts to queue.

Browse .docx Files

Files: 0 / 2 | Size: 0.00 MB | Est. tokens: 0 in / 0 out | Est. cost: Free

Cost basis: Paid (~$0.13/M in · $0.38/M out) — via OpenRouter (google/gemma-4-31b-it).

Methodology: ~10% of DOCX bytes are translatable text · ~4 bytes/token · 1.5× input headroom for coverage retry · output ≈ 70% of input. Real cost will typically land at 50–100% of this estimate; safe upper bound for budget planning.