Noemad Docs
Model catalog
Discovery, capabilities, official pricing sources, and model specifications.
Integration guide · planned capabilities. Model-specific capabilities, usage rates, and x402 upto settlement require backend implementation and validation. The preview runs local samples; documented contracts do not imply live availability.
Implementation target. This guide specifies the intended product surface. The current frontend uses local samples; it does not call an inference service or collect funds. An endpoint is enabled only after its adapter, ownership, usage, and payment behavior are qualified.
Authenticated discovery is authoritative
GET /v1/models returns the models available to the product key or supported owner session. A public model name does not prove provider entitlement, quota, health, or an enabled Noemad route. Only qualified models may be presented as runnable.
GET /v1/models
GET /v1/models?type=embedding
GET /v1/models?type=rerank
GET /calculate/models/calculate/models describes quoteable pricing specifications. A fresh quote validates request-specific options and creates an immutable cost snapshot. The browser must not maintain a competing pricing engine for real payments.
Capability fields
| Field | Meaning |
|---|---|
| id / label / vendor | Stable public identifier, display name, and model developer |
| modality / input_modalities | Endpoint family and accepted input kinds |
| context_length / max_output | Published limits; the admitted request may be bounded more tightly |
| thinking | Supported reasoning values, including auto or none only when declared |
| native_dimensions / dimensions | Embedding shape and any supported selectable dimensions |
| max_documents | Rerank candidate bound |
| pricing unit / minimum / modifier | How this exact configuration is measured |
| enabled / health / entitlement | Separate live deployment facts; absent data is unknown |
Complete archived model reference
Model specification reference. Names and capability limits come from the imported catalog snapshot. Enabled models require fresh qualification. USD rates below are listed only when verified for the exact model against an official source; unverified prices are marked.
The library preserves 63 catalog records plus 9 additional entries from the exported capability pages: two later chat entries, two video entries, three speech entries, realtime, and typed evaluation. Browse and filter the entire collection in the model library.
| Family | Reference records | Integration guide |
|---|---|---|
| Chat | 42 | Chat & text |
| Embeddings | 12 | Embeddings |
| Rerank | 5 | Rerank |
| Image | 6 | Media |
| Video | 2 | Durable video jobs |
| Speech synthesis / transcription | 3 | Audio |
| Realtime | 1 | Realtime |
| Structured evaluation | 1 | Typed decisions |
Chat capability records
| Model reference | Input | Context / output | Reasoning |
|---|---|---|---|
GLM 5.3glm-5.3 | text | 1,000,000 / 128,000 | auto, low, medium, high, xhigh, max |
GLM 5.3 Flashglm-5.3-flash | text, image, video | 1,000,000 / 128,000 | auto, low, medium, high, xhigh, max |
GLM 5.2glm-5.2 | text | 1,000,000 / 128,000 | auto, low, medium, high, xhigh, max |
GLM 5.1glm-5.1 | text | 200,000 / 128,000 | auto, low, medium, high |
GLM 5.0 Turboglm-5.0-turbo | text | 200,000 / 128,000 | auto, low, medium, high |
GLM 5V Turboglm-5v-turbo | text, image, video | 200,000 / 128,000 | auto, low, medium, high |
Muse Spark 1.1muse-spark-1.1 | text | 1,048,576 / 1,000,000 | auto, none, low, medium, high, xhigh, max |
Muse Spark 1.2muse-spark-1.2 | text | 1,048,576 / 1,000,000 | auto, none, low, medium, high, xhigh, max |
Qwen 3.8 Maxqwen3.8-max | text, image, video | 1,000,000 / 64,000 | auto, none, low, medium, high |
Qwen 3.7 Maxqwen3.7-max | text | 1,000,000 / 64,000 | auto, none, low, medium, high |
Qwen 3.7 Plusqwen3.7-plus | text, image | 1,000,000 / 64,000 | auto, none, low, medium, high |
Kimi K3kimi-k3 | text, image, video | 1,000,000 / 64,000 | auto, low, high, max |
Kimi K2.7kimi-k2.7 | text, image | 256,000 / 32,000 | auto, low, medium, high |
Kimi K2.6kimi-k2.6 | text, image | 256,000 / 32,000 | auto, low, medium, high |
Kimi K2.5kimi-k2.5 | text, image | 256,000 / 32,000 | auto, low, medium, high |
DeepSeek V4 Prodeepseek-v4-pro | text, image | 1,000,000 / 384,000 | auto, low, medium, high |
DeepSeek V4 Pro 0813deepseek-v4-pro-0813 | text | 1,000,000 / 384,000 | auto, none, low, medium, high, max |
DeepSeek V4 Flashdeepseek-v4-flash | text | 1,000,000 / 384,000 | auto, low, medium, high |
DeepSeek V4.1 Flashdeepseek-v4.1-flash | text, image | 1,000,000 / 128,000 | auto, none, low, medium, high, max |
DeepSeek V4 Flash 0731deepseek-v4-flash-0731 | text | 1,000,000 / 384,000 | auto, none, low, medium, high, max |
MiniMax M3minimax-m3 | text, image, video | 1,000,000 / 48,000 | auto, low, medium, high |
MiniMax M2.7minimax-m2.7 | text, image | 200,000 / 48,000 | auto, low, medium, high |
Hunyuan 3hy3 | text | 1,000,000 / 64,000 | auto, low, medium, high |
Hunyuan 4hy4 | text | 1,000,000 / 64,000 | auto, low, medium, high |
Grok 4.5grok-4.5 | text, image, file | 500,000 / 64,000 | auto, minimal, low, medium, high, xhigh |
Grok 4.6grok-4.6 | text, image, file | 500,000 / 64,000 | auto, low, medium, high, xhigh |
GPT 5.5gpt-5.5 | text, image, file | 1,050,000 / 128,000 | auto, none, low, medium, high, xhigh |
GPT 6 Astragpt-6-astra | text, image, video, file | 1,050,000 / 128,000 | auto, low, medium, high, xhigh, max |
GPT 5.6 Solgpt-5.6-sol | text, image, file | 1,050,000 / 128,000 | auto, none, low, medium, high, xhigh, max |
GPT 5.6 Terragpt-5.6-terra | text, image, file | 1,050,000 / 128,000 | auto, none, low, medium, high, xhigh, max |
GPT 5.6 Lunagpt-5.6-luna | text, image, file | 1,050,000 / 128,000 | auto, none, low, medium, high, xhigh, max |
Claude Fable 5.1claude-fable-5.1 | text, image, file | 1,000,000 / 128,000 | auto, low, medium, high, xhigh, max |
Claude Fable 5claude-fable-5 | text, image, file | 1,000,000 / 128,000 | auto, low, medium, high, xhigh, max |
Claude Opus 5claude-opus-5 | text, image, file | 1,000,000 / 128,000 | auto, low, medium, high, xhigh, max |
Claude Opus 4.8claude-opus-4.8 | text, image, file | 1,000,000 / 128,000 | auto, low, medium, high, xhigh, max |
Claude Opus 4.7claude-opus-4.7 | text, image, file | 1,000,000 / 128,000 | auto, low, medium, high, xhigh, max |
Claude Opus 4.6claude-opus-4.6 | text, image, file | 1,000,000 / 128,000 | auto, none, low, medium, high, xhigh, max |
Claude Sonnet 5claude-sonnet-5 | text, image, file | 1,000,000 / 128,000 | auto, low, medium, high, xhigh, max |
Claude Sonnet 4.6claude-sonnet-4.6 | text, image, file | 1,000,000 / 128,000 | auto, none, low, medium, high, xhigh, max |
Claude Haiku 4.5claude-haiku-4.5 | text, image, file | 200,000 / 64,000 | auto, low, medium, high, xhigh, max |
GLM 5.3 FlashXglm-5.3-flashx | text, image, video | 1,000,000 / 128,000 | Not specified |
Grok 4.7grok-4.7 | text, image, file | 500,000 / 64,000 | Not specified |
Specialized records
| Model reference | Capability | Native dimension / bound |
|---|---|---|
Gpt 5.5 Imagegpt-5.5-image | Images | Adapter-specific |
Gpt 5.6 Luna Imagegpt-5.6-luna-image | Images | Adapter-specific |
Gpt 5.6 Sol Imagegpt-5.6-sol-image | Images | Adapter-specific |
Gpt 5.6 Terra Imagegpt-5.6-terra-image | Images | Adapter-specific |
Grok Imagine Imagegrok-imagine-image | Images | Adapter-specific |
Grok Imagine Image 2.0grok-imagine-image-2.0 | Images | Adapter-specific |
BGE Small ENbge-small-en | Embeddings | 384 dimensions |
BGE Base ENbge-base-en | Embeddings | 768 dimensions |
BGE Large ENbge-large-en | Embeddings | 1024 dimensions |
BGE-M3bge-m3 | Embeddings | 1024 dimensions |
EmbeddingGemma 300Membeddinggemma-300m | Embeddings | 768 dimensions |
Qwen3 Embedding 0.6Bqwen3-embedding-0.6b | Embeddings | 1024 dimensions |
BGE Reranker Basebge-reranker-base | Rerank | 100 documents |
voyage-4-litevoyage-4-lite | Embeddings | 1024 dimensions |
voyage-4voyage-4 | Embeddings | 1024 dimensions |
voyage-4-largevoyage-4-large | Embeddings | 1024 dimensions |
voyage-code-4voyage-code-4 | Embeddings | 1024 dimensions |
voyage-finance-2voyage-finance-2 | Embeddings | 1024 dimensions |
voyage-law-2voyage-law-2 | Embeddings | 1024 dimensions |
Voyage rerank-3voyage-rerank-3 | Rerank | 100 documents |
Voyage rerank-3-litevoyage-rerank-3-lite | Rerank | 100 documents |
Voyage rerank-2.5voyage-rerank-2.5 | Rerank | 100 documents |
Voyage rerank-2.5-litevoyage-rerank-2.5-lite | Rerank | 100 documents |
Grok Imagine Videogrok-imagine-video | Video | Adapter-specific |
Grok Imagine Video 1.5grok-imagine-video-1.5 | Video | Adapter-specific |
Grok Voice TTSgrok-voice-tts | Speech synthesis | Adapter-specific |
Grok Voice Transcribe 1.0grok-voice-transcribe-1.0 | Transcription | Adapter-specific |
Grok Voice Transcribe 2.0grok-voice-transcribe-2.0 | Transcription | Adapter-specific |
GPT Livegpt-live | Realtime | Adapter-specific |
Jev 1.13jev-1.13 | Structured evaluation | Adapter-specific |
Changes are explicit
A model alias, provider mapping, supported tool, rate, or maximum changes the request contract. Publish new configuration for future quotes and preserve accepted snapshots. If a provider cannot honor a pinned rate, the operator absorbs the difference or rejects before dispatch; it must not increase the user’s approved charge.