Today's price · per 1M tokens
Input
$0.600
per 1M tokens
Output
$3.60
per 1M tokens
Blended
$1.35
blended $/1M — 3:1 weighted input:output
Cached input
$0.060
10% of input — prompt caching
How this price is scoped: Requests with more than 200K input tokens are billed at twice the listed prices for the whole request, on input, cached input and output alike.
Price receipt
We read Apodex's own pricing page on and found Apodex 1.1 listed with a price on the same row — that read is where the number above comes from, recorded as a price increase. We keep the page text we read; its content hash is 6e228db5f4.
Overview
Apodex's full 397B reasoning model, built for hard reasoning, long-context and quality-critical research work, and the model behind Apodex's own research workbench. Text only, 262,144-token context; it thinks before answering and bills those reasoning tokens as output. Served first-party over OpenAI-compatible and Anthropic Messages endpoints at the same price. Priced at $0.60 input / $0.06 cached / $3.60 output per 1M tokens.
Capabilities
struck through = not supportedBenchmark performance
accuracy % · higher is betterWe hold no per-benchmark accuracy scores for this model yet, so it has no accuracy average. That is a gap in our coverage, not a sign the model is untested — independent evaluators often publish a composite index for a new model long before releasing its per-benchmark numbers. The percentile above is its standing across the independent composites below.
- AA Intelligence Index: 13.7 (Artificial Analysis composite across reasoning, knowledge and coding evals)
- AA-Omniscience Index: -27.9 (−100–100)
- GDPval-AA v2: 1194.8 ELO (Real-world work tasks, human baseline = 1000)
- HLE-Diamond: 11.6% (measured 2026-11-22) — A revised 1,000-question set of Humanity's Last Exam, half reasoning, half knowledge
- CritPt v2, hard subset: 17% (measured 2026-11-22) — The hardest research-level physics problems in CritPt v2, solved with a code sandbox
- FrontierCode-AA 2.0: 6.9% (measured 2026-11-22) — Coding tasks in real open-source repositories, graded on whether a maintainer would merge the change
- Terminal-Bench-Science: 0.5% (measured 2026-11-22) — Expert-written science research workflows completed in a terminal
- GDP.pdf: 11.2% (measured 2026-11-22) — Tasks over long professional documents; share of tasks that pass every criterion
- AutomationBench-AA: 28.2% (measured 2026-11-22) — Multi-step work across simulated business apps, without breaking the rules set
- GPQA Diamond: 86.4% (measured 2026-11-22)
- AA-LCR (long-context reasoning): 79.3% (measured 2026-11-22)
- Terminal-Bench 2.1: 69.7% (measured 2026-11-22)
- Terminal-Bench 4.0: 0% (measured 2026-11-22)
- SciCode: 45.5% (measured 2026-11-22)
- MMMU-Pro: 79.2% (measured 2026-11-22)
- Cost to run the AA Intelligence Index: $0.764 per task
- Ranks #139 of 207 comparably-measured models by percentile score, across 3 independent measurements
- Ranks #140 of 207 comparably-tested models by normalized performance per dollar
Source: · updated · See full rankings →
Specifications
- Provider
- Apodex
- Context window
- 262K tokens
- Modality
- text
- Parameters
- 397B
- Open source
- No — proprietary
- Released
- Status
- Current
- Last updated
- Tags
Availability verified: — listed on Apodex's own page