AI Battle > Rankings > Arena Preference Elo

Arena Preference Elo Ranking — Speech-to-Speech (2026)

null by null leads the Arena Preference Elo ranking with a score of 1096. The ranking covers 20 evaluated speech-to-speech from a public dataset that AI Battle re-checks daily.

Last updated: . Source: public benchmark datasets and provider pricing, aggregated daily; 20 entries shown.

Gemini 3.1 Flash Live Minimal1096Gemini 3.1 Flash Live High1063GPT-Live-1 (Sol, low)1053GPT-Live-1 (Astra, medium)1048Cartesia Line (Default Cascaded System)1027Grok Voice Think Fast 2.0 High1011GPT-Realtime-1.51000ElevenLabs Agents (Default Cascaded System)993GPT Realtime (Aug '25)974Nova 2.0 Sonic (Mar 2026)974GPT-Realtime-2.1 Minimal937GPT-Realtime-2 (High)932GPT-Realtime-2.1 High928GPT Realtime Mini (Oct '25)917GPT-Realtime-2 (Minimal)916Deepgram Voice Agent (Default Cascaded Syste914Nova Sonic913Grok Voice Think Fast 1.0908Inworld Realtime (Default Cascaded System)867Qwen3.5 Omni Plus Realtime859
Every entry in the dataset, scaled from the smallest to the largest published value.

All 20 entries in this dataset

#EntryCreatorValueΔ vs #1
1Gemini 3.1 Flash Live Minimal10960.0%
2Gemini 3.1 Flash Live High1063-3.0%
3GPT-Live-1 (Sol, low)1053-3.9%
4GPT-Live-1 (Astra, medium)1048-4.4%
5Cartesia Line (Default Cascaded System)1027-6.3%
6Grok Voice Think Fast 2.0 High1011-7.8%
7GPT-Realtime-1.51000-8.7%
8ElevenLabs Agents (Default Cascaded System)993-9.3%
9GPT Realtime (Aug '25)974-11.1%
10Nova 2.0 Sonic (Mar 2026)974-11.1%
11GPT-Realtime-2.1 Minimal937-14.5%
12GPT-Realtime-2 (High)932-15.0%
13GPT-Realtime-2.1 High928-15.3%
14GPT Realtime Mini (Oct '25)917-16.3%
15GPT-Realtime-2 (Minimal)916-16.4%
16Deepgram Voice Agent (Default Cascaded System)914-16.6%
17Nova Sonic913-16.7%
18Grok Voice Think Fast 1.0908-17.1%
19Inworld Realtime (Default Cascaded System)867-20.9%
20Qwen3.5 Omni Plus Realtime859-21.6%

How the values are distributed

Across the 20 numeric entries in this window the median is 956 and the mean is 967, so the typical entry sits 140 below the leader. The window spans 859 to 1096.

25322231
8 equal buckets between the smallest (859) and largest (1096) value; numbers under bars are entry counts.

10 of 20 entries are at or above the median, and the top three hold 17% of the combined value — read it against the metric below, since a higher number is not "better" for every dataset.

Who publishes these entries

The source dataset for this family does not publish a provider or creator label per entry, so AI Battle cannot group them: the table above is the complete published list. Where the dataset does carry a label (model, gateway and tool families do), this section names the creators and how many entries each holds.

Metric reference

This ranking carries the metrics elo_lower, elo_mid, elo_upper from the Arena Preference Elo dataset in the tab::s2s family. Units and the direction of "better" are not declared by the source dataset, so AI Battle does not invent them: values are shown exactly as published, and the ordering is the source dataset's own.

How this ranking is built

Rebuilt on from the public dataset Arena Preference Elo (family tab::s2s). The source lists 20 entries and all 20 are shown here. AI Battle requests the dataset's rows, keeps the ordering the source publishes, and computes only the statistics on this page: median, mean, spread, the #1–#2 gap, the creator breakdown and the raw-data file. Nothing on this page is generated text about the entries — the numbers come from the source.

Refresh cadence: every deploy re-pulls the dataset, so this page tracks the source's own update cycle. Corrections: see the methodology page.

Limitations

Raw data

Machine-readable copy of this table, with the statistics below and the source provenance: /data/rankings/s2s-arena-preference-elo.json. Free to reuse with attribution to AI Battle and the source dataset.

FAQ — Arena Preference Elo

What is the best model for Arena Preference Elo in 2026?

null by null currently leads the Arena Preference Elo ranking with a score of 1096, the highest value in this dataset across Speech-to-Speech. AI Battle aggregates the dataset from its public source and re-checks it daily.

How often is the Arena Preference Elo ranking updated?

The Arena Preference Elo ranking is rebuilt daily from the upstream public dataset. Scores are the source dataset's own published values, so positions track whatever that source last released — each ranking page names the dataset it comes from.

Which models rank in the top 3 for Arena Preference Elo?

The top 3 are: 1. null (1096), 2. null (1063), 3. null (1053).

All AI Battle Rankings

Directories