Understand the numbers
Ranges
The leaderboard gives you a position. The benchmark also publishes an interval around it. At rank 40, that interval covers 24 different models — and they cost between $0.280 and $50.00 per million tokens. MiMo-V2.6-Flash: Price under confirmation since Oct 12, 2026.
These bands are wider than the truth, on purpose. They are built from each model’s own published interval and nothing else, which ignores that the scores are jointly estimated from shared battles — a covariance the published intervals cannot recover. So a band shows what is not ruled out, never what is identified, and two models sharing a position are not thereby tied or equivalent.
Stand in a position
Pick a rank. These are the models whose published interval covers it.
24 models are not ruled out of rank 40. The cheapest bills $0.280 per million output tokens and the dearest $50.00 — a spread of 178.6× at one position. MiMo-V2.6-Flash: Price under confirmation since Oct 12, 2026.
| Model | Not ruled out of | $/M out |
|---|---|---|
| MiMo-V2.6-Flash | 22–41 | $0.280Price under confirmation since Oct 12, 2026 |
| GPT-5.4 | 26–41 | $15.00 |
| Claude Opus 4.5 | 28–41 | $25.00 |
| Grok 4.5 | 29–46 | $6.00 |
| GPT-5.6 Terra | 30–48 | $12.00 |
| GPT-6.1 Sol | 26–54 | $10.00 |
| GLM 5 | 31–48 | $2.24 |
| DeepSeek V4 Pro 0423 | 29–54 | $3.48Price under confirmation since Oct 12, 2026 |
| Kimi K2.5 | 31–48 | $3.00 |
| Gemma 4 31B | 31–57 | $0.400 |
| Claude Sonnet 5 | 34–54 | $10.00 |
| Step 5 Preview | 31–60 | $2.70 |
| Inkling | 34–54 | $4.05 |
| Qwen3.8 27B | 34–57 | $1.98Price under confirmation since Oct 12, 2026 |
| GPT-6 Astra | 34–59 | $50.00 |
| GLM 4.6 | 35–57 | $1.75 |
| Hy3 | 34–61 | $0.528 |
| Claude Sonnet 4.5 | 38–57 | $15.00 |
| Qwen3.5 397B A17B | 38–58 | $2.34 |
| GLM 5V Turbo | 35–62 | $4.00 |
| Qwen3.6 Plus | 38–61 | $1.95 |
| GLM 4.7 | 38–62 | $1.75Price under confirmation since Oct 12, 2026 |
| Gemini 3.5 Flash Lite | 38–62 | $2.50 |
| Gemma 4 26B A4B | 38–66 | $0.300Price under confirmation since Oct 12, 2026 |
The widest position on this board is rank 62: 24 models, spanning 417× in price.
Every rated model, drawn as the positions its interval does not rule out
This is not a claim that the ranking is wrong. A rank column is a deterministic display of a sort over point estimates, and the benchmark never claimed statistical separation. It is also the only board in this market that publishes an interval at all, which is the only reason this page can exist.
206 models in the catalogue carry no entry on this board. They are absent from the figure and named as absent — never drawn at rank zero, and never counted as scoring badly. Absence of a measurement is not a measurement.
Try it yourself
The blind rank test shows two models with their names hidden and asks you to pick the higher-scoring one — then tells you whether the published intervals can tell them apart at all.
Scores and intervals are LMArena Elo, used under CC BY 4.0 from the official dataset. Prices are list output price per million tokens, as published by each provider.
Questions this page answers
Does this mean the leaderboard is wrong?
No. A rank column is a deterministic display of a sort over point estimates, and the benchmark never claimed statistical separation. It is also the only board in this market that publishes an interval for every model, which is the only reason this page can be drawn.
Are two models sharing a position tied?
No, and this page will not say so. Failure to rule a position out is not evidence of equivalence. The bands are built from each model's own published interval, which ignores that scores are jointly estimated from shared battles, so they are wider than the truth.
Why are some models missing from the figure?
A model is drawn only if it carries an entry on this board and a paid price. Models with no entry are named as absent rather than drawn at rank zero. Absence of a measurement is not a measurement, and it is never a verdict.
Where do the numbers come from?
Scores and intervals are LMArena Elo under CC BY 4.0 from the official dataset. Prices are list output price per million tokens as published by each provider. Everything on this page is arithmetic on those two.
Read citation text and BibTeX
410 of 413 adjacent published positions have overlapping score intervals. The figure covers 143 models with measured rank bands and a paid output-token price. Intervals overlapping a position do not establish equivalence or task success. Prices are USD per million output tokens. As of 2026-10-13. Lens: LMArena Elo (CC BY 4.0). https://undominated.ai/ranges/ Method: https://undominated.ai/methodology/ Arena data: https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (CC BY 4.0). Derived calculations by Undominated.ai. Publication: sha256:7f43bf1ba022be9bb0d9d5462e7815de12aa148353b947efac4756084700b179; calculation SHA-256: 2be71186e159799b8a18bc0ae29953e6e0b7ab863bc824228f9f7201f785c716; permitted inputs: https://undominated.ai/data/publications/manifests/7f43bf1ba022be9bb0d9d5462e7815de12aa148353b947efac4756084700b179.json
@misc{undominated-ranges-2026-10-13,
author = {{Undominated.ai}},
title = {ranges evidence — Undominated.ai},
year = {2026},
url = {https://undominated.ai/ranges/},
note = {as of 2026-10-13; LMArena Elo (CC BY 4.0)},
annote = {Publication: sha256:7f43bf1ba022be9bb0d9d5462e7815de12aa148353b947efac4756084700b179; calculation SHA-256: 2be71186e159799b8a18bc0ae29953e6e0b7ab863bc824228f9f7201f785c716; permitted inputs: https://undominated.ai/data/publications/manifests/7f43bf1ba022be9bb0d9d5462e7815de12aa148353b947efac4756084700b179.json}
}