Nearly at the level of the best models in the world.

Two independent leaderboards, recorded this day. We publish no Numezis scores: we publish what third parties measure.

01 · Intelligence index4.8 %

The gap between the best open model and the best closed model, across 127 models measured on the same 22 benchmarks. Artificial Analysis, 29.08.2026.

02 · Coding arena1 %

The gap between the best open model (Kimi K3, 1 674) and the best closed model (Claude Opus 5, 1 691), on anonymous human votes. LMArena, 29.08.2026.

03 · Best installable15th / 127

The rank of the most powerful model an organisation can actually run inside its walls — GLM-5.3-Flash. Ahead of half the models on the market.

04 · Cost per task×26

What a task costs on a top-tier API ($2.34) versus an installed open model ($0.09). Artificial Analysis, 29.08.2026.

What third parties measure.

Every rank is verifiable on the source page, at the date of the snapshot. Classes change every quarter; what does not change is the order of magnitude.

Coding arena · LMArena · 29.08.2026
ModelScoreType
1Claude Opus 5 (max)API1 691
2Kimi K3 (max)OPEN1 674
3Qwen 3.8 MaxOPEN1 669
4Claude Opus 5 (high)API1 663
5GLM-5.3 (max)OPEN1 599
6Qwen 3.8 27BOPEN1 595
7Gemini 3.7 Flash (high)API1 587
Intelligence index · Artificial Analysis · 29.08.2026
ModelRankScoreMachine
Qwen 3.8 27BOPEN30of 127 models52N1 · N2 · N4
Qwen 3.8 Flash-NextOPEN20of 127 models56N2 · N4
GLM-5.3-FlashOPEN15of 127 models57N2 · N4
DeepSeek V4 FlashOPEN31of 127 models52N4
Claude Opus 5 (high)API4of 127 models61
Claude Opus 5 (max)API1of 127 models63

The “max” versions that top the leaderboards (Kimi K3: 2.8 trillion parameters, Qwen 3.8 Max, GLM-5.3) need clusters of several dozen cards and remain API services. They serve here as a level reference. What the machine installs is the catalogue above — the most powerful models an organisation can actually run inside its own walls.

Sources: LMArena ↗ · Artificial Analysis ↗ · Deloitte ↗recorded 29.08.2026

Power is not what costs.

An installed open model has a marginal cost tending to zero. The same work, on API, is paid per task.

The calculator
01 · Per task$0.09

The cost of a task on GLM-5.3-Flash, installed at your site. The top-tier API bills $2.34 for the same task.

02 · Per year70'000 – 134'000 CHF

The API bill avoided for an organisation of 20 on regular use — 73'000 tasks a year. The calculator runs the curve with your numbers.

03 · Over three years50%

Of inference costs saved on-site versus public cloud, at significant volume. Deloitte, State of AI in the Enterprise.

Not promises. Named models.

Verified on their official cards, 29 August 2026: size, memory, license. This is what is installed, not the API versions of the leaderboards.

ModelParametersMemory (BF16)MachineLicense
Qwen 3.8 27BAlibaba · 256 k tokens27 B · dense≈ 56 Go (BF16)N1 · N2 · N4Apache 2.0
GLM-5.3-FlashZ.ai · 256 k tokens320 B · 18 B actifs≈ 320 Go (BF16)N2 · N4MIT
DeepSeek V4 FlashDeepSeek · 1 M tokens284 B · 13 B actifs≈ 291 Go (BF16)N4MIT
Apertus 70BSWISSEPFL / ETH / CSCS · 131 k tokens71 B · dense≈ 142 Go (BF16)N2 · N4Apache 2.0
Apertus v1.5 8BSWISSEPFL / ETH / CSCS · 131 k tokens9 B · dense≈ 18 Go (BF16)N1 · N2 · N4Apache 2.0

Memory: size of the model weights in BF16, excluding context memory. Switching between models is included in the service, every quarter. Apertus is the Swiss model (EPFL / ETH / CSCS), proposed by default for the public and para-public sector.

100% applies only to the architecture.

A benchmark measures what a model can do. It does not measure where your documents are while it does it.

Talk to an engineer
01 · Your dataNever leaves

The fundamental difference: a 97% model running in your building beats a 100% model running on someone else's.

02 · What is 100%The absence of flow

No outbound flow of client content, by construction — not by promise. Certified annually, producible to your auditor.

03 · What is never 100%Any system

No architecture is free of all risk. The difference is knowing what one can prove — and producing it, before one's professional body.