ATM-Bench
Agents answer questions about a person’s history using images, videos, and email records. The Harbor package contains 31 hard questions with schema-guided memory.
The README agent table available November 19, 2026 reports GPT-5.5 at extra-high effort first with 48.1%, followed by Claude Opus 4.7 at 46.6%.
- Made by
- Jingbiao Mei and colleagues
- Released
- March 2026
- Kind
- Finding facts in long inputs
- Level
- University
- Scored
- Checked automatically
- Answers
- Short answer
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Agentic
- Environment
- APIs and tools
- Items from
- Real work and use
- Refresh
- One fixed set
- Run by
- Anyone
- Strongest model tested
- GPT-5.4 #28 on the capability index
- Last model added
- June 2026
- Scores refreshed
- November 20th, 2026
Claude Opus 4.5 leads with 86% among models released by June 2026 Claude Opus 4.5 leads with 86% among models released by June 2026
Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran on its own with several settings, we keep its best. Agents other companies built around a model get a pale shade of that model's lab color. Gray marks blends and entries not in our model list, such as fine-tunes and older models. None of these count toward leaders. Agents built around a model, such as Graphon V1 + GPT-6 Astra, are their own entries. We don't count them as the model's score. This view shows only models we track, each at its best.
The newest model with a score here came out in June 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 86.0 Claude Opus 4.5
- 85.3 GPT-5
- 85.1 GPT-5.4
- 84.5
Qwen3.6 27B
- 84.1 GPT-5.2
- 83.6
MiniMax-M3
- 83.3 Claude Sonnet 4.5
- 82.6 MiMo-V2.5
- 81.7 Graphon V1 + GPT-6 Astra
- 79.2 Graphon V1 + DeepSeek V4.1 Flash
- 78.6
Gemini 2.5 Pro
- 78.2
Qwen3-VL-8B-Instruct
- 76.9
Gemini 2.5 Flash
- 76.7 DeepSeek V4.1
- 71.7 MiMo-V2.5-Pro
- 68.4 DeepSeek V4 Pro (OpenRouter)
- 68.0 Memexa + DeepSeek-V4-Flash
- 56.8 MemPalace + Qwen3-VL-8B-Instruct
- 56.1 Qwen3-VL-8B-Instruct (OpenRouter)
- 52.5 ScrapMem (No-Forget) + Qwen3-VL-8B-Instruct
- 51.0 ATM-RAG + Qwen3-VL-8B-Instruct
- 50.3 Self-RAG + Qwen3-VL-8B-Instruct
- 48.4 ScrapMem + Qwen3-VL-8B-Instruct
- 47.2 MemoryOS + Qwen3-VL-8B-Instruct
- 44.8 A-Mem + Qwen3-VL-8B-Instruct
- 43.5 Mem0 + Qwen3-VL-8B-Instruct
- 42.9 HippoRAG2 + Qwen3-VL-8B-Instruct
- 34.8
Qwen3-VL-2B-Instruct
- 27.3 SimpleMem + Qwen3-VL-8B-Instruct
- 1 Claude Opus 4.5 Oracle 86%
- 2 GPT-5 Oracle 85.3%
- 3 GPT-5.4 Oracle 85.1%
- 4
Qwen3.6 27B Oracle 84.5%
- 5 GPT-5.2 Oracle 84.1%
- 6
MiniMax-M3 Oracle 83.6%
- 7 Claude Sonnet 4.5 Oracle 83.3%
- 8 MiMo-V2.5 Oracle 82.6%
- 9 Graphon V1 + GPT-6 Astra Memory / Graphon V1 / Graphon V1 81.7%
- 10 Graphon V1 + DeepSeek V4.1 Flash Memory / Graphon V1 / Graphon V1 79.2%
- 11
Gemini 2.5 Pro Jun 2025, Oracle 78.6%
- 12
Qwen3-VL-8B-Instruct Oracle 78.2%
- 13
Gemini 2.5 Flash Jun 2025, Oracle 76.9%
- 14 DeepSeek V4.1 Memory / DeepSeek V4.1 / BGE-M3 + bge-reranker-v2-m3 76.7%
- 15 MiMo-V2.5-Pro Oracle 71.7%
- 16 DeepSeek V4 Pro (OpenRouter) RAG / Qwen3-VL-2B-Instruct / Voyage-4-large + Voyage rerank-2.5 68.4%
- 17 Memexa + DeepSeek-V4-Flash Memory / DeepSeek-V4-flash + Qwen3.6-27B captions / BGE-M3 + bge-reranker-v2-m3 68%
- 18 MemPalace + Qwen3-VL-8B-Instruct Memory / Qwen3-VL-2B-Instruct / MiniLM-L6 56.8%
- 19 Qwen3-VL-8B-Instruct (OpenRouter) RAG / Qwen3-VL-2B-Instruct / MiniLM-L6 + BM25 56.1%
- 20 ScrapMem (No-Forget) + Qwen3-VL-8B-Instruct Memory / Qwen3-VL-2B-Instruct / MiniLM-L6 52.5%
- 21 ATM-RAG + Qwen3-VL-8B-Instruct RAG / Qwen3-VL-2B-Instruct / MiniLM-L6 51%
- 22 Self-RAG + Qwen3-VL-8B-Instruct RAG / Qwen3-VL-2B-Instruct / MiniLM-L6 50.3%
- 23 ScrapMem + Qwen3-VL-8B-Instruct Memory / Qwen3-VL-2B-Instruct / MiniLM-L6 48.4%
- 24 MemoryOS + Qwen3-VL-8B-Instruct Memory / Qwen3-VL-2B-Instruct / MiniLM-L6 47.2%
- 25 A-Mem + Qwen3-VL-8B-Instruct Memory / Qwen3-VL-2B-Instruct / MiniLM-L6 44.8%
- 26 Mem0 + Qwen3-VL-8B-Instruct Memory / Qwen3-VL-2B-Instruct / MiniLM-L6 43.5%
- 27 HippoRAG2 + Qwen3-VL-8B-Instruct RAG / Qwen3-VL-2B-Instruct / MiniLM-L6 42.9%
- 28
Qwen3-VL-2B-Instruct Oracle 34.8%
- 29 SimpleMem + Qwen3-VL-8B-Instruct Memory / Qwen3-VL-2B-Instruct / MiniLM-L6 27.3%
Source: Official ATM-Bench leaderboard
- 1 Claude Opus 4.5 Oracle 86%
- 2 GPT-5 Oracle 85.3%
- 3 GPT-5.4 Oracle 85.1%
- 4
Qwen3.6 27B Oracle 84.5%
- 5 GPT-5.2 Oracle 84.1%
- 6
MiniMax-M3 Oracle 83.6%
- 7 Claude Sonnet 4.5 Oracle 83.3%
- 8 MiMo-V2.5 Oracle 82.6%
- 9
Gemini 2.5 Pro Jun 2025, Oracle 78.6%
- 10
Qwen3-VL-8B-Instruct Oracle 78.2%
- 11
Gemini 2.5 Flash Jun 2025, Oracle 76.9%
- 12 MiMo-V2.5-Pro Oracle 71.7%
- 13
Qwen3-VL-2B-Instruct Oracle 34.8%
Source: Official ATM-Bench leaderboard
| # | Entry | Score | Setting | Date | Link | Source |
|---|---|---|---|---|---|---|
| 1 | Claude Opus 4.5 | 86% | Oracle | Nov 2025 | Official leaderboard | |
| 2 | GPT-5 | 85.3% | Oracle | Aug 2025 | Official leaderboard | |
| 3 | GPT-5.4 | 85.1% | Oracle | Mar 2026 | Official leaderboard | |
| 4 | | 84.5% | Oracle | Apr 2026 | Official leaderboard | |
| 5 | GPT-5.2 | 84.1% | Oracle | Dec 2025 | Official leaderboard | |
| 6 | | 83.6% | Oracle | Jun 2026 | Official leaderboard | |
| 7 | Claude Sonnet 4.5 | 83.3% | Oracle | Sep 2025 | Official leaderboard | |
| 8 | MiMo-V2.5 | 82.6% | Oracle | Apr 2026 | Official leaderboard | |
| 9 | Graphon V1 + GPT-6 Astra | 81.7% | Memory / Graphon V1 / Graphon V1 | Sep 2026 | graphon.ai | Official leaderboard |
| 10 | Graphon V1 + DeepSeek V4.1 Flash | 79.2% | Memory / Graphon V1 / Graphon V1 | Sep 2026 | graphon.ai | Official leaderboard |
| 11 | | 78.6% | Oracle | Jun 2025 | Official leaderboard | |
| 12 | | 78.2% | Oracle | Oct 2025 | Official leaderboard | |
| 13 | | 76.9% | Oracle | Jun 2025 | Official leaderboard | |
| 14 | DeepSeek V4.1 | 76.7% | Memory / DeepSeek V4.1 / BGE-M3 + bge-reranker-v2-m3, Memexa v2 | github.com | Official leaderboard | |
| 15 | MiMo-V2.5-Pro | 71.7% | Oracle | Apr 2026 | Official leaderboard | |
| 16 | DeepSeek V4 Pro (OpenRouter) | 68.4% | RAG / Qwen3-VL-2B-Instruct / Voyage-4-large + Voyage rerank-2.5, RE-call | github.com | Official leaderboard | |
| 17 | Memexa + DeepSeek-V4-Flash | 68% | Memory / DeepSeek-V4-flash + Qwen3.6-27B captions / BGE-M3 + bge-reranker-v2-m3 | Apr 2026 | github.com | Official leaderboard |
| 18 | MemPalace + Qwen3-VL-8B-Instruct | 56.8% | Memory / Qwen3-VL-2B-Instruct / MiniLM-L6 | Oct 2025 | github.com | Official leaderboard |
| 19 | Qwen3-VL-8B-Instruct (OpenRouter) | 56.1% | RAG / Qwen3-VL-2B-Instruct / MiniLM-L6 + BM25, MediaFind | mediafind.io | Official leaderboard | |
| 20 | ScrapMem (No-Forget) + Qwen3-VL-8B-Instruct | 52.5% | Memory / Qwen3-VL-2B-Instruct / MiniLM-L6 | Oct 2025 | arxiv.org | Official leaderboard |
| 21 | ATM-RAG + Qwen3-VL-8B-Instruct | 51% | RAG / Qwen3-VL-2B-Instruct / MiniLM-L6 | Oct 2025 | github.com | Official leaderboard |
| 22 | Self-RAG + Qwen3-VL-8B-Instruct | 50.3% | RAG / Qwen3-VL-2B-Instruct / MiniLM-L6 | Oct 2025 | github.com | Official leaderboard |
| 23 | ScrapMem + Qwen3-VL-8B-Instruct | 48.4% | Memory / Qwen3-VL-2B-Instruct / MiniLM-L6 | Oct 2025 | arxiv.org | Official leaderboard |
| 24 | MemoryOS + Qwen3-VL-8B-Instruct | 47.2% | Memory / Qwen3-VL-2B-Instruct / MiniLM-L6 | Oct 2025 | github.com | Official leaderboard |
| 25 | A-Mem + Qwen3-VL-8B-Instruct | 44.8% | Memory / Qwen3-VL-2B-Instruct / MiniLM-L6 | Oct 2025 | github.com | Official leaderboard |
| 26 | Mem0 + Qwen3-VL-8B-Instruct | 43.5% | Memory / Qwen3-VL-2B-Instruct / MiniLM-L6 | Oct 2025 | github.com | Official leaderboard |
| 27 | HippoRAG2 + Qwen3-VL-8B-Instruct | 42.9% | RAG / Qwen3-VL-2B-Instruct / MiniLM-L6 | Oct 2025 | github.com | Official leaderboard |
| 28 | | 34.8% | Oracle | Oct 2025 | Official leaderboard | |
| 29 | SimpleMem + Qwen3-VL-8B-Instruct | 27.3% | Memory / Qwen3-VL-2B-Instruct / MiniLM-L6 | Oct 2025 | github.com | Official leaderboard |
| # | Model | Score | Setting | Date | Link | Source |
|---|---|---|---|---|---|---|
| 1 | Claude Opus 4.5 | 86% | Oracle | Nov 2025 | Official leaderboard | |
| 2 | GPT-5 | 85.3% | Oracle | Aug 2025 | Official leaderboard | |
| 3 | GPT-5.4 | 85.1% | Oracle | Mar 2026 | Official leaderboard | |
| 4 | | 84.5% | Oracle | Apr 2026 | Official leaderboard | |
| 5 | GPT-5.2 | 84.1% | Oracle | Dec 2025 | Official leaderboard | |
| 6 | | 83.6% | Oracle | Jun 2026 | Official leaderboard | |
| 7 | Claude Sonnet 4.5 | 83.3% | Oracle | Sep 2025 | Official leaderboard | |
| 8 | MiMo-V2.5 | 82.6% | Oracle | Apr 2026 | Official leaderboard | |
| 9 | | 78.6% | Oracle | Jun 2025 | Official leaderboard | |
| 10 | | 78.2% | Oracle | Oct 2025 | Official leaderboard | |
| 11 | | 76.9% | Oracle | Jun 2025 | Official leaderboard | |
| 12 | MiMo-V2.5-Pro | 71.7% | Oracle | Apr 2026 | Official leaderboard | |
| 13 | | 34.8% | Oracle | Oct 2025 | Official leaderboard |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official ATM-Bench leaderboard
GPT-5 on the official leaderboard passed 85% in August 2025
Models released before it came out already scored that high. All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official ATM-Bench 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 Long context
- MRCR v2 In use Multi-round coreference: retrieve the n-th of several similar requests hidden in a long synthetic conversation, up to 1M tokens.
Muse Spark 1.3 98.5%
- LongBench v2 In use 503 multiple-choice questions over 8K-2M-word contexts (QA, dialogue, code repos, structured data). Accuracy.
Inkling 2 Small 70.6%
- GraphWalks In use A graph is given as an edge list in long context. The model answers BFS or parent queries, scored by F1. Claude Fable 5 99.96%
- OpenAI MRCR v2 In use Models read a long synthetic chat that hides 2, 4 or 8 identical writing requests and must reproduce the response to the i-th one. Scored by mean string match ratio per context bin.
Gemini 3.7 Flash 89%
- CL-bench In use Model must learn new knowledge from a supplied long context (rules, procedures, data) and solve tasks. Strict LLM-verified rubrics. GPT-5.4 27.9%
- AA-LCR 1.1 In use Models answer 100 questions that need reasoning across sets of real documents of about 100K tokens. An LLM equality checker with a fixed system prompt compares each answer with the key.
Step 5 Preview 88.3%
Data audit
All 9 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
- Pass Independent runs Scores from the official leaderboard (29 entries).
- Pass Models scored 13 models we track, plus 16 other entries on its leaderboard.
- Pass Newest score The newest is from June 2026, for MiniMax-M3.
- Note Frontier models None of the top 10 has a score, which is fine for a retired test. The strongest tested is GPT-5.4 (#28).
- Note Headroom Saturated. The best is 86%, past 85% since August 2025.
Leaderboard import 3 pass
- Pass Importer We read the board on atmbench.github.io: 27 rows for models we track and 3 other entries at the last read.
- Pass Last read November 20th, 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.