Search docs

Search docs, endpoints, errors

Pricing

Simple usage-based pricing. Pay per token — no subscriptions, no minimums.

Core Models (Apodex 1.1 and 1.1 Mini) are free to call for the duration of the campaign — no balance and no coupon code required. List prices resume automatically when it ends.

Struck-through prices are the standard list prices. The discounted prices apply automatically at settlement for the duration of the campaign — no code or coupon needed.

Two product lines, two billing models: core models are single-pass inference billed on real tokens; deep research tiers run an autonomous agent task billed on the tokens it consumes plus per-use tools. See the core models vs. deep research comparison.

LLM API — Core Models

Direct single-model inference over /v1/chat/completions and /v1/responses. You pay exactly for the tokens you send and receive.

ModelInput / 1M tokensCached inputOutput / 1M tokensAccess
Apodex 1.1Free
apodex-1.1
The 397B flagship. Highest reasoning and instruction-following quality across long documents and complex tasks.
$0.30$0.00$0.03$0.00$3.00$0.00Available to all
Apodex 1.1 MiniFree
apodex-1.1-mini
The 35B sibling. Much lower latency and cost, tuned for high-volume classification, extraction and drafting.
$0.10$0.00$0.01$0.00$1.00$0.00Available to all

Core models (apodex-1.1, apodex-1.1-mini) are billed purely on real tokens, in three tiers: input, cached input (cheaper, served from the prefix cache) and output. Tokens already generated are billed even if the client disconnects mid-stream.

Core model requests with more than 200K input tokens are billed at 2× the listed prices for the entire request — the multiplier applies to input, cached input and output alike.

Deep Research API

Agentic research tiers: one request launches a task that plans, searches and iterates — typically for minutes, consuming tokens well beyond the final report, plus hosted tool calls.

ModelInput / 1M tokensOutput / 1M tokensAccess
Deep Research1.1Limited-time -20%
apodex-1-1-deep-research
Fast, source-grounded research that finds, compares, and synthesizes knowledge into clear, actionable intelligence.
$5.00$4.00$20.00$16.00Available to all
Deep Solve1.1Limited-time -20%
apodex-1-1-deep-solve
Resolve what matters.
$5.00$4.00$25.00$20.00Available to all
Deep Discover1.1PreviewLimited-time -20%
apodex-1-1-deep-discover
Highest-rigor solver for the hardest, frontier problems — explores paths, tests hypotheses, verifies claims, and converges on the strongest answer.
$10.00$8.00$100.00$80.00Request Access
Deep Research1.0
apodex-1-0-deep-research
Previous-generation Apodex 1.0 tier — still available at its original list price.
$10.00$40.00Available to all
Deep Solve1.0
apodex-1-0-deep-solve
Previous-generation Apodex 1.0 tier — still available at its original list price.
$10.00$50.00Available to all
Deep Discover1.0Preview
apodex-1-0-deep-discover
Previous-generation Apodex 1.0 tier — still available at its original list price.
$10.00$100.00Request Access

Prices are list prices in USD per 1 million tokens. Input counts your initial request (including history and files); output covers every token the agent produces while working — reasoning and intermediate steps included, not just the final answer.

Hosted tool calls on the deep research tiers are billed per use, on top of tokens: web search $5 per 1,000 calls, web fetch $6 per 1,000 calls, code sandbox $90 per 1,000 calls.

Tiers marked Preview (currently Deep Discover) require access through the Apodex Frontier Program — approved accounts can call them like any other model. All other tiers are generally available.

Promotional credits, when granted, are applied automatically before your paid balance.

Docs | Apodex API Platform