Catalog

gpt-6-astra

Frontier reasoning model for complex, long-context, and agentic workloads.

gpt-6.1-sol

Reasoning model for coding, professional work, and agentic workflows.

gpt-6-sol

Balanced reasoning model for coding and agentic workflows.

gpt-6-luna

Efficient reasoning model for focused, high-volume tasks.

gpt-5.6-sol

Flagship general-purpose model for complex reasoning, coding, and long-context work.

gpt-5.6-terra

Balanced general-purpose model for reasoning, coding, and production workloads.

gpt-5.6-luna

Efficient general-purpose model for fast, high-volume reasoning and chat.

gpt-5.5

High-capability general-purpose model for reasoning, chat, and developer workflows.

gpt-5-mini

Earlier small reasoning model for existing integrations and focused tasks.

gpt-5.3-codex

Code-focused reasoning model for agentic software-development workflows.

gpt-5.4-mini

Smaller, faster general-purpose model for everyday production tasks.

gpt-5.4-nano

Lightweight model for high-volume tasks where speed and unit cost matter.

gpt-image-2.5-flare

Latest-generation image model for high-fidelity generation and editing from text and image references.

gpt-image-2.5-sunburst

Latest-generation image model for high-fidelity generation and editing from text and image references.

gpt-image-2

Generate or edit images from text and image references.

gpt-image-1-mini

Generate images efficiently from text and supported references.

sora-2

Generate short video clips from text prompts through an asynchronous job.

gpt-transcribe

Latest-generation speech-to-text with per-audio-minute billing.

gpt-4o-transcribe

Transcribe audio into text with token-level measured usage.

gpt-4o-transcribe-diarize

Transcribe audio with speaker-aware diarized segments.

gpt-4o-mini-transcribe

Fast, economical audio transcription for production speech workloads.

gpt-4o-mini-tts

Coming soon

Generate spoken audio from text through the Primex audio API.

text-embedding-3-large

Create high-quality vector embeddings for retrieval and semantic search.

text-embedding-3-small

Create efficient vector embeddings for retrieval, search, and indexing.

text-embedding-ada-002

Create legacy-compatible vector embeddings for integrations that require ada-002.

whisper

Transcribe general-purpose audio into text with audio-minute billing.

gpt-5.4

Unavailable

High-capability model for reasoning, chat, and professional workflows.

gpt-4o

Familiar multimodal model for existing OpenAI integrations.

gpt-5

Original GPT-5 reasoning model for existing integrations.

gpt-4.1

Instruction-following non-reasoning model for existing integrations.

claude-fable-5-1

Claude model for long-context reasoning and general assistant workloads.

claude-opus-5-5

High-capability Claude model for demanding reasoning, analysis, and agentic work.

claude-fable-5

Claude model for long-context reasoning and general assistant workloads.

claude-opus-5

High-capability Claude model for demanding reasoning, analysis, and agentic work.

claude-sonnet-5-5

Claude model for reasoning, coding, and long-running agents.

claude-sonnet-5

Balanced Claude model for reasoning, coding, and production assistants.

claude-haiku-4-5

Fast, economical Claude model for responsive and high-volume tasks.

claude-sonnet-4-6

Balanced Claude model for reasoning, coding, and production assistants.

claude-opus-4-8

High-capability Claude model for demanding reasoning, analysis, and agentic work.

claude-haiku-5-5

Fast reasoning for focused tasks, extraction, and coding.

grok-4.6

Flagship reasoning model for analysis, chat, and tool-oriented workloads.

grok-4.3

General-purpose reasoning model for analysis, chat, and tool-oriented workloads.

grok-4-1-fast-reasoning

Coming soon

Low-cost fast model with extended reasoning for latency-sensitive workloads.

grok-4-1-fast-non-reasoning

Coming soon

Low-cost fast model for quick chat and tool-oriented workloads.

grok-4.7

Multimodal reasoning, coding, and tool-oriented workloads.

DeepSeek-V4-Pro

High-capability reasoning model for complex analysis and production assistants.

DeepSeek-V4-Flash

Faster, lower-cost reasoning model for latency-sensitive workloads.

DeepSeek-V3.2

Coming soon

Efficient V3-line reasoning model for general analysis and chat.

DeepSeek-V3.2-Speciale

Specialized V3.2 reasoning variant for demanding analysis.

Llama-4-Maverick

General-purpose mixture-of-experts model for chat and long-context workloads.

Llama-4-Scout

Efficient mixture-of-experts model for chat and long-context workloads.

Llama-3.3-70B-Instruct

Dense multilingual instruct model for general chat and reasoning.

model-router

Selects a suitable available model for each request while preserving one Primex API surface.

Phi-4

Compact Phi text model for existing integrations.

Phi-4-mini-instruct

Economical Phi text model for focused tasks.

Mistral-Large-3

General-purpose Mistral model for reasoning, chat, and multilingual work.

mistral-medium-3-5

Capable mid-tier Mistral model for reasoning, chat, and multilingual work.

Codestral-2501

Code-specialized model for generation, explanation, and repository tasks.

Kimi-K2.7-Code

Long-context code model for software-development and agentic coding tasks.

Kimi-K2.6

Long-context general-purpose model for reasoning and document-heavy conversations.

eleven-v4

Expressive, high-fidelity speech across 90+ languages.

eleven-v4-turbo

Low-latency expressive speech across 90+ languages.

eleven-v3

Expressive speech for dramatic performance and dialogue.

eleven-v3-conversational

Expressive, low-latency conversational speech.

eleven-multilingual-v2

Generate multilingual speech with Eleven Multilingual v2 through the Primex audio API.

eleven-flash-v2-5

Fast multilingual speech across 32 languages.

eleven-flash-v2

Fast English-only speech.

cohere-command-a

Enterprise chat and reasoning model for retrieval-grounded workflows.

embed-v-4-0

Create multilingual vector embeddings for retrieval and semantic search.

Cohere-embed-v3-multilingual

Unavailable

Create multilingual vector embeddings for retrieval and semantic search.

Cohere-command-a-plus-05-2026

Coming soon

Newer, efficient Cohere Command model for enterprise chat and reasoning.

Cohere-Embed-V5-Fast

Fast multilingual text vectors for search queries and retrieval.

Cohere-Embed-V5-Pro

High-quality multilingual text vectors for document indexing and retrieval.

FLUX.2-pro

Generate high-fidelity images from text prompts with megapixel-based billing.

FLUX.2-flex

Coming soon

Generate images from text prompts with adjustable quality and flat megapixel-based billing.

GLM-5.3-Flash

GLM 5.3 Flash: economical text chat and reasoning.

GLM-5.3

GLM 5.3: text reasoning, coding and long-horizon tasks.

GLM-5.2-Fast

GLM 5.2 Fast: text chat, coding and agent workflows.

Nemotron-Lightning-3.5

Coming soon

Nemotron Lightning 3.5: economical text chat and reasoning.

Nemotron-3-Ultra

Nemotron 3 Ultra: text reasoning and long-context agent workflows.

MiniMax-M3

MiniMax M3: economical text chat and reasoning.