MemLens (32K) - Temporal Reasoning: leaderboard
Accuracy in percent on the MemLens 194 questions about dates, durations and the order of events, at a 32K-token interleaved image and text conversation history, with the evidence sessions hidden among distractor sessions (MemLens, 2026), judged by Qwen3-VL-235B. Higher is better.
Metric: Accuracy (%) on the temporal reasoning (TR, 194 questions) at a 32K-token history, interleaved multimodal multi-session conversation history with the evidence sessions hidden among haystack sessions, judged correct or incorrect by a Qwen3-VL-235B-A22B-Instruct judge; higher is better. Source: arxiv.org. Saturation forecast: Not forecast. 27 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 VL 30B A3B Instruct | 60.82 | #895 |
| 2 | Qwen 3 VL 8B Instruct | 58.25 | #1029 |
| 3 | Qwen 3 VL 235B A22B Instruct | 55.67 | #532 |
| 4 | Kimi K2.5 | 53.09 | #266 |
| 5 | GLM-4.6V | 53.09 | #710 |
| 6 | Qwen 3 VL 8B (Thinking) | 52.58 | #910 |
| 7 | Qwen 3.5 122B A10B | 51.55 | #346 |
| 8 | Qwen 3 VL 2B Instruct | 50.52 | #2367 |
| 9 | Qwen 3 VL 235B A22B (Thinking) | 49.48 | #459 |
| 10 | Cosmos-Reason2-8B | 48.45 | #755 |
| 11 | Nemotron Nano 12B v2 VL | 44.85 | #2368 |
| 12 | Gemma 3 12B (IT) | 43.81 | #1887 |
| 13 | Phi-4 Multimodal Instruct | 43.81 | #4154 |
| 14 | Qwen 3 VL 4B (Thinking) | 43.3 | #1334 |
| 15 | Qwen 3 VL 4B Instruct | 42.78 | #1332 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=memlens-32k-temporal-reasoning · How It Works · Data refreshed daily, snapshot 2026-11-21.