Benchmark Almanac

X-Stream

Gemini 3 Pro 49.6 leads since June 2026
Not in model cards

Models answer questions that combine concurrent video streams. A model judge grades answers, while forward-looking questions also require responses within the correct time window.

The benchmark includes 4,220 question-answer pairs across 932 videos. It reports early responses and missing responses separately.

Released
June 2026
Kind
Video understanding
Level
Everyday
Scored
Graded by a model
Answers
Short answer
Tasks
All public
Measures
Capability
Language
English
In and out
Text, Video → Text
Interaction
Single turn
Refresh
One fixed set
Run by
Anyone
In views
Vision
Strongest model tested
Gemini 3 Pro #56 on the capability index
Last model added
November 2025
Scores refreshed
November 18th, 2026

In the authors’ words

While video streaming understanding has made significant strides, real-world applications, such as live sports broadcasting, autonomous driving, and multi-screen collaboration, inherently demand continuous, multi-stream interactions. However, existing benchmarks are confined to single-stream paradigms, leaving a critical gap in evaluating online, cross-stream reasoning. To bridge this, we introduce X-Stream, the first benchmark dedicated to multi-stream streaming understanding. Comprising 4,220 rigorously curated QA pairs across 932 videos, X-Stream evaluates 11 subtasks across multi-window, multi-view, and multi-device scenarios. Crucially, our dataset is constructed using a novel dual-verification pipeline that prevents over-reliance on a single stream. Furthermore, we pioneer the conceptualization of multi-modal large language models (MLLMs) as naive multiplexers, systematically evaluating their performance through the lens of Signal Multiplexing Theory. Our extensive online inference experiments reveal a stark reality: state-of-the-art MLLMs struggle significantly with concurrent streams, achieving only about 50% score and exhibiting poor proactive ability. Ultimately, X-Stream exposes the trade-off of current multiplexing schemes, providing both a practical evaluation protocol and empirical guidance for next-generation multi-stream agents.

Abstract of X-Stream: Exploring MLLMs as Multiplexers for Multi-Stream Understanding (arXiv:2606.02482, CC0)
12
Entries in its paper, 5 models we track

Gemini 3 Pro leads with 49.6 among models released by November 2025 Gemini 3 Pro leads with 49.6 among models released by November 2025

Best score per entry from the benchmark's paper, read on November 18th, 2026. A paper’s table doesn’t add newer models. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.

The newest model with a score here came out in November 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 Gemini 3 Pro 49.6
  2. 2 Doubao-Seed-1.8 36.8
  3. 3 Qwen3-Omni-30B-A3B 34.3
  4. 4 Qwen3-VL-30B-A3B 34.2
  5. 5 GPT-5 27.8
  6. 6 Qwen2.5-Omni-7B 26.8
  7. 7 Qwen3-VL-8B 26.8
  8. 8 Qwen2.5-VL-7B 25.5
  9. 9 GPT-4o (snapshot unspecified) 22.5
  10. 10 Dispider 15.4
  11. 11 VideoLLM-online-8B 8.48
  12. 12 MMDuet2 6.79

Source: The X-Stream paper

#EntryScoreSettingDateSource
1 Gemini 3 Pro 49.6 Nov 2025 Official leaderboard
2 Doubao-Seed-1.8 36.8 Official leaderboard
3 Qwen3-Omni-30B-A3B 34.3 Sep 2025 Official leaderboard
4 Qwen3-VL-30B-A3B 34.2 Official leaderboard
5 GPT-5 27.8 Aug 2025 Official leaderboard
6 Qwen2.5-Omni-7B 26.8 Mar 2025 Official leaderboard
7 Qwen3-VL-8B 26.8 Official leaderboard
8 Qwen2.5-VL-7B 25.5 Jan 2025 Official leaderboard
9 GPT-4o (snapshot unspecified) 22.5 Official leaderboard
10 Dispider 15.4 Official leaderboard
11 VideoLLM-online-8B 8.48 Official leaderboard
12 MMDuet2 6.79 Official leaderboard
#ModelScoreSettingDateSource
1 Gemini 3 Pro 49.6 Nov 2025 Official leaderboard
2 Qwen3-Omni-30B-A3B 34.3 Sep 2025 Official leaderboard
3 GPT-5 27.8 Aug 2025 Official leaderboard
4 Qwen2.5-Omni-7B 26.8 Mar 2025 Official leaderboard
5 Qwen2.5-VL-7B 25.5 Jan 2025 Official leaderboard

Rankings by part

We read 4 parts from the benchmark’s paper, each ranking models on one part of the test, such as Instant, Backward and Forward. Gemini 3 Pro leads every one of them. Arrows show how far a model moved from its overall rank among the models we track.

Instant

5 models we track have a score here.

  1. 1 Gemini 3 Pro 73.4
  2. 2 Qwen3-Omni-30B-A3B 63.9
  3. 3 GPT-5 44.3
  4. 4 Qwen2.5-Omni-7B 42.0
  5. 5 Qwen2.5-VL-7B 40.0

Number one over time

Lab claims count from the day the card came out, independent runs from the model’s release.

Google DeepMind
Benchmark paper Not tested since Jun 2026

Source: The X-Stream paper

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

Plot with other benchmarks
In the benchmark’s paper
25.0 30.0 35.0 40.0 45.0 50.0 2025 2026 25.0 30.0 35.0 40.0 45.0 50.0 2025 2026

Source: Official X-Stream leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

More in Audio and video

All 695 in Audio and video
  • Video-MMMU In use 300 expert-level lecture videos with questions testing perception, comprehension and adaptation of knowledge. Accuracy. Qwen 3.8 Max 88.7%
  • LVBench In use Hours-long video understanding with multiple-choice questions on reasoning, grounding and summarization. Gemini 4 Argon 91.7%
  • MMVU In use 3,000 expert-annotated questions on videos across 27 disciplines. Multiple-choice and open-ended accuracy. Qwen 3.8 Max 82.4%
  • MLVU In use Multi-task long video understanding (minutes to hours) with multiple-choice and generation tasks. Accuracy. Qwen 3.8 Max 90.8%
  • FLEURS In use Models transcribe read speech in 102 languages built from FLoRes-101 sentences. Model cards usually report word error rate, so lower scores are better. Phi-4-multimodal 4.00%
  • VoiceBench In use Spoken instructions with speaker, environment and content variation. Voice assistant responses scored by GPT judge or rules. Qwen3.5-Omni-Plus 93.1%

Data audit

8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 12 months ago, in November 2025, for Gemini 3 Pro. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
  • Pass Independent runs Scores from the benchmark’s paper (12 entries).
  • Pass Models scored 5 models we track, plus 7 other entries on its leaderboard.
  • Warning Newest score The newest came 12 months ago, in November 2025, for Gemini 3 Pro.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Gemini 3 Pro (#56).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on arxiv.org: 5 rows for models we track and 7 other entries at the last read.
  • Pass Last read November 18th, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check