Skip to main content
Radar

Five or fewer confirmed AI changes, with original sources, on mornings when something changed.A free source-linked morning brief.

Start the free Radar Brief
BenchLM

Frontier model index · August 2026

Frontier AI Models

Kimi K3 is #3 overall with a verified frontier position. The frontier is the current top 10 in each ranking view. Exact-source coverage decides whether a position is verified or stays provisional.

See the current frontier Read the scoring method
Frontier size
10 models
Verified frontier
5 models
Candidates
5 models
Leader to cutoff
13 pts

Live index

Which models are on the frontier now?

The table changes with the selected capability. Rank comes from the provisional lane; the status badge shows whether exact-source evidence also supports the position.

Cutoff: 73

#ModelScore

Method

Where we draw the line

Frontier membership is deliberately narrow: the ten highest provisional scores in the selected view. Verification is a second gate. A model needs at least 8 exact-source benchmark rows, plus eligible coverage across two or more categories, before its position moves into the verified lane.

Missing rows are not filled in. External consensus can support a provisional rank, but it cannot grant verified status.

  1. 01

    Rank

    Order eligible models by the selected score.

  2. 02

    Cut

    Keep the first ten rows.

  3. 03

    Verify

    Apply exact-source coverage gates.

Cost check

The price of frontier performance

The top score and the cheapest acceptable score are often different buying decisions. This view keeps the evidence status attached to both.

Open the full price chart

Upper-left is better

Scores use the selected frontier view. Prices are official output-token rates in USD per million tokens.

Lowest priced frontier API

Claude Sonnet 5

$10 / 1M out

Claude Mythos 5Claude Opus 5Kimi K3GPT-5.6 SolClaude Fable 5Claude Opus 4.8GPT-5.6 TerraClaude Sonnet 5

Baseline ledger

What changed this month

Qwen3.8 Max

Released this month and currently #6 in the verified frontier.

Verified

Common questions

Frontier model questions, answered from the data

Which are the frontier models right now?

As of August 2026, the frontier is the overall top 10 on this page: Claude Mythos 5, Claude Opus 5, and Kimi K3 lead the ranking, and 5 of the 10 hold verified exact-source coverage. The full list above updates as scores and evidence change.

What is the difference between frontier models and foundation models?

Foundation model describes a class: large models trained broadly so they can be adapted to many tasks. Frontier model describes a rank: the small, shifting subset of foundation models at the top of current measured capability. Every frontier model is a foundation model; most foundation models are not frontier. Likewise, LLM names the architecture family — frontier names where a model currently stands.

Why are they called frontier models?

The frontier is the outer edge of demonstrated AI capability — a moving boundary, not a fixed specification. A model is on the frontier only relative to everything else available at the time, which is why this index is dated, re-scored as benchmark evidence lands, and records when models enter or leave.

How big are frontier models?

Mostly undisclosed. Closed labs stopped publishing parameter counts, and recent open-weight frontier entries that do publish them are mixture-of-experts designs with totals in the hundreds of billions of parameters. Size is a weak proxy for capability, so this index tracks what is measurable instead: benchmark scores with exact sources, and what the model costs to run.