Model List
The full catalog — every model, its exact id, current prices, context length and capabilities — lives in the console:
console.meganova.ai/serverless
It is generated from the same records that bill your requests, so the prices there are the prices you pay. This page describes what is in the catalog and how to navigate it; it does not restate it, because a hand-maintained copy drifts from the live rates the moment a model is repriced.
Finding a model
The catalog page carries filters for the things people usually pick on:
Search
A model by name or id
Provider — Claude, Gemini, DeepSeek, Qwen, GLM, OpenAI, Kimi, MiniMax, Mistral, Llama, Manta, …
Everything from one vendor
Task — Text Generation, Multimodal, Vision, Text To Image, Text To Video, Image To Video, Text To Speech, Speech To Text, Embeddings, Reranker, Realtime
Models for one kind of work
Free
Models callable without a deposit
Show Enterprise only
Models that also run on dedicated enterprise deployments
Each model's page shows its id (with a copy control), input/output and cached-input prices, context length, per-tier daily limits, and a playground to try it before writing code.
What the catalog contains
Frontier models from Anthropic (Claude Opus, Sonnet, Haiku), OpenAI (GPT-5 family), Google (Gemini Pro, Flash, Flash-Lite), xAI (Grok), and Alibaba (Qwen Plus, Qwen-VL) — for work where quality matters more than unit cost.
Open-weight models from DeepSeek, Z.ai (GLM), Moonshot (Kimi), MiniMax, Xiaomi (MiMo), Mistral, Meta (Llama) and Google (Gemma) — strong reasoning and coding at a fraction of frontier prices.
Manta, MegaNova's own adaptive line (Mini, Flash, Pro), which routes each request to a model sized for it. See Introducing Manta.
Media and utility models — image generation, text-to-video and image-to-video, text-to-speech and speech-to-text, embeddings, and reranking.
Community fine-tunes for creative writing and character dialogue. The catalog groups these under their own category chip.
Free models
A subset is callable with no deposit, subject to a daily request limit that grows with your tier. Filter by Free in the catalog for the current set, and see Free Model Quota for how the limits and the paid fallback work.
Programmatic access
The same catalog is available over the API, which is the right source if you are building a model picker rather than reading one:
Returns OpenAI-shaped model objects including pricing, context length and capability flags. On the Enterprise Endpoint, the same call returns only the models served from dedicated deployments.
See Also
Core Services — quickstart and the per-modality guides
Free Model Quota — daily limits and the paid fallback
Pricing Tiers — what each tier unlocks
Enterprise Endpoint — the same models on dedicated deployments
Last updated