Choose a text-to-video model for MCP workflows

Choose between the two current Studio text-to-video entries

Start with the live `list_models` response, filter for `text-to-video`, and use the returned model ID rather than guessing an identifier from a display name. In the current public Studio catalog, the two text-to-video entries are Kling 2.6 Pro at 42 credits, labeled for social video, and Veo 3.1 Fast at 96 credits, labeled for premium campaigns. Those labels are Studio positioning, not evidence that one model has better quality, prompt adherence, reliability, or creator preference.

Treat the 42- and 96-credit figures as budget fields to present, not a recommendation to start with Kling or escalate to Veo. A model choice needs a separate task-specific decision and runtime evidence. Recheck the catalog immediately before any spend because this page follows a monthly product-freshness cadence and no authenticated model response was observed during research.

The disclosed fal endpoints document different direct-provider envelopes. Kling documents five- or ten-second video in 16:9, 9:16, or 1:1. Veo documents four-, six-, or eight-second video in 16:9 or 9:16. OfflineCreator's public MCP `generate` schema accepts `modelId`, `prompt`, optional `aspectRatio`, and optional `wait`; it does not expose duration, resolution, audio, negative-prompt, or guidance controls. Provider options therefore must not be copied into an MCP call or described as Studio controls.

Kling 2.6 Pro

Studio: text-to-video · 42 credits · social videoDirect fal documentation: 5 or 10 seconds; 16:9, 9:16, or 1:1. Studio MCP does not publish a duration input.

Veo 3.1 Fast

Studio: text-to-video · 96 credits · premium campaignsDirect fal documentation: 4, 6, or 8 seconds; 16:9 or 9:16. Studio MCP does not publish duration, resolution, or audio inputs.

Build a catalog-discovery record before choosing a model

Capture the live discovery response before drafting an executable request: returned ID, display name, workflow, displayed credits, retrieval time, and any model-specific fields the connected service actually supplies. This page can name the two public catalog entries, but it cannot fill in their Studio IDs, durations, or supported ratios from the sources reviewed.

Next record the creative requirement separately: text-to-video workflow, intended placement, maximum approved spend, and whether the brief aligns with a catalog positioning label. Do not convert “social video” or “premium campaigns” into a quality ranking. Send a true Kling-versus-Veo evaluation to the dedicated pairwise page rather than duplicating that comparison here.

Finally, compare every proposed argument with the connected MCP schema. The direct fal pages expose duration and audio-related controls that the public Studio MCP schema does not. Leave unsupported fields unresolved, require model-specific evidence for any ratio, and let the operator see the current charge before a later task-specific guide prepares generation.

Record missing compatibility fields instead of inventing a matrix

The public catalog establishes two text-to-video names, workflows, positioning labels, and credit costs. The package establishes a shared request shape. Neither source supplies the exact current Studio IDs, a per-model ratio matrix, selectable duration, resolution, audio behavior, or successful runtime observation. A useful inventory marks each of those fields `unverified` rather than borrowing values from fal.

Supported statements on Choose a text-to-video model for MCP workflows stay limited to documented OfflineCreator Router behavior and the cited primary sources. This article does not invent unpublished routes, benchmarks, privacy promises, or customer outcomes.

The inventory is complete only as a dated public-catalog snapshot, not as an execution matrix. If the connected response differs from the public snapshot, preserve the response and route the decision through the maintained parent directory. Do not infer availability, compatibility, or quality from an entry merely appearing on a marketing page.

Unverified runtime fields

Exact Studio IDs, model-specific ratios, Studio durations, successful completionDirect-provider documentation and the shared MCP enum do not fill these catalog gaps.

Take the next action from the decision you made

This article answers “mcp text to video models” for OfflineCreator Studio. It complements the AI Router overview and does not replace model how-tos, unpublished routes, or outcome claims.

What this article covers

This article explains the narrow query `mcp text to video models`: discover and inventory the current Studio text-to-video entries using names, workflow, catalog positioning, credit charge, and the public MCP request boundary. The dedicated `/models/kling-vs-veo` page owns pairwise comparison design and evidence limits. This page does not own provider API integration, image-to-video selection, an executable model recommendation, runtime promises, or claims that a more expensive generation produces a better result.

This article does not treat community or social retrieval as proof of OfflineCreator quality, pricing, or customer results. Product statements stay limited to documented Studio routing, quotes, and provider disclosure for mcp text to video models.

The authenticated Studio model response was not queried, no production credential was accessed, and no paid output was generated. Current product facts were rechecked against public Studio pages and the public MCP implementation; direct model envelopes were rechecked against fal documentation. The draft remains unpublished and unreviewed, with a gap for a reproducible matched contact sheet. If the catalog fields cannot be maintained or the proof requirement becomes mandatory for publication, consolidate the decision into the parent model directory.