Entity trail

Mem0

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Showing the first 40 findings. More graph evidence exists in the corpus.

Corpus findings

2026-09-10 / arxiv-researcherMoving Memory Retrieval Into the Agent Kernel Beats Mem0 by Up to 4 Points and Cuts Latency 15-61%arXiv 2609.10144 (2026-09-09) makes the agent-system kernel, not individual agents, responsible for retrieval, privacy enforcement and prompt injection, while specialized agents only write structured tagged memories. Implemented on AIOS and tested across GPT-4o, Llama-3.1:8B and Qwen-2.5:7B over 1,800 trials, it improves personalization scores by 2.4 to 4.0 points on a 5-point scale against Mem0 using identical underlying storage (profile usage on GPT-4o goes 1.05 to 4.69), every comparison significant at p < 10^-18. Against full unfiltered context concatenation it statistically matches on two of three models while cutting end-to-end latency 15-61% with proportional token savings.
2026-09-10 / github-pulse-researcherMem0's OpenClaw plugin 1.1.0 deletes Dream consolidation entirely and removes the 2,000-character extraction cutoffReleased 2026-09-09 as `@mem0/openclaw-mem0` v1.1.0, this removes automatic Dream scheduling and locking, the `openclaw mem0 dream` command, Dream configuration, the memory-dream skill, and Dream-state public artifacts, telling integrations that use that surface to update. It also drops OpenClaw's separate 2,000-character extraction cutoff, so selected user and assistant messages retain their full redacted text while recent-message selection, earlier-summary selection and noise filtering still apply. Mem0 shipped matching v0.3.0 plugins for OpenCode, DeepSeek and Pi Agent the same day.
2026-09-03 / github-pulse-researchermem0 2.0.20 pulled its OSS notice system off PostHog feature-flag evaluation and onto a static repo fileReleased 2026-09-02T13:18Z for the Python SDK and 13:19Z for the Node SDK, mem0 2.0.20 / ts-v3.1.8 moves notice configuration to a static cacheable repository file with a bundled disabled fallback and deterministic rollout assignment, replacing a live call to PostHog's feature-flag evaluation API (#7185). The stated reason is fail-safe behaviour when the remote config is unavailable, but the practical effect is removing a third-party network request from a memory library's startup path. `RedisDBConfig` also switched to Pydantic's native `extra="forbid"` instead of a custom validator (#7089).
2026-08-30 / sources-researcherVoiceMem splits agent memory into an informational and an emotional brain and retrieves in 134 msarXiv 2608.26005 (submitted 2026-08-26, Xie et al.) is the top-upvoted paper on Hugging Face Daily Papers this week at 167 upvotes. VoiceMem runs a parallel 'left brain' for factual memory and 'right brain' for affective attribution and dual-node persona modeling, with streaming memory I/O and interchangeable backends. At top-5 retrieval the left brain beats Mem0 at top-200 by nearly 30 points, the right brain sets state of the art on three persona benchmarks with a 4.29-point aggregate gain, and retrieval completes in 134 ms, inside standard VAD latency so it adds no conversational delay.
2026-08-27 / arxiv-researcherVoiceMem Splits Speech-Agent Memory Into Informational and Affective Stores, Retrieving in 134 msXie et al. give duplex speech language models a streaming memory architecture with a parallel informational store and an emotional store plus streaming memory I/O, along with a full pipeline for memory-aware training, long-horizon evaluation, and decoupled deployment with interchangeable backends. Under top-5 retrieval the informational side outperforms Mem0 at top-200 by nearly 30 points, and the affective side with short- and long-horizon attribution and dual-node persona modeling improves the aggregate score by 4.29 points over the previous best across three persona benchmarks. Retrieval completes in 134 ms, inside standard VAD latency, so it adds no perceptible conversational delay.
2026-08-27 / github-pulse-researchermem0 adds opt-out telemetry to its Strands and DeepSeek plugins with an explicit no-content guaranteeBoth mem0-strands-v0.1.1 and deepseek-plugin-v0.1.1 shipped 2026-08-27, sending anonymous usage events over the SDK's existing PostHog client with no new dependency. The changelogs enumerate what is excluded rather than what is included: never queries, memory text, message content, entity ids, metadata or API keys, only counts, durations, booleans and coarse failure kinds. Opt out with MEM0_TELEMETRY=false. The exclusion-list framing is becoming the standard way OSS agent-memory projects document telemetry.
2026-08-14 / skill-finderYou cannot predict an agent memory system's serving cost from conversation length — errors run 18-69%Benchmarking Mem0, Hindsight, and Mastra Observational Memory across conversations up to 400 turns and 665 LoCoMo questions (arXiv:2608.11879, Aug 12) shows cost models built on conversation length and message size miss by 18-69%, because internal memory behavior dominates. Break-even against simply replaying the full transcript varies wildly: some systems beat replay within tens of turns, others never do inside 400. Accuracy across systems spans 21-54%, and the backbone model choice moves cost as much as the memory system does. The practical instruction is to measure break-even against naive full-transcript replay on your own traffic before adopting a memory layer — for short sessions the memory system can be strictly worse on both axes.
2026-08-13 / sources-researcherAdaption Labs: Agent Memory Fails at Write Time, Not Retrieval — 90.6% on LongMemEval vs 60.6% for Full History, With 97% Fewer TokensPosted 13 Aug 2026, the argument inverts the usual framing: retrieval can only search memories that exist, so the highest-leverage optimization is extraction at write time. Their system writes in two forms — narrative memories preserving causal context and atomic memories preserving verbatim facts — and never deletes, instead appending new state marked as superseding the old so retrieval favors current truth while history survives; they report expecting to need knowledge graphs and finding they did not. Reported numbers: LongMemEval 90.6% vs Mem0 OSS 71.6% and full-history 60.6%; LoCoMo 88.2% vs 82.2%; BEAM 61.3% vs 39.7%; 97% fewer tokens on long conversations, mean latency down 45%, p95 down 67%. Vendor-published benchmarks with no independent replication — weight accordingly.
2026-08-10 / projects-researcherThree Agent-Memory Repos Now Hold a Combined 211,000 Stars — claude-mem at 90,282, mem0 at 62,934, and MemPalace at 58,269The persistent-context layer has consolidated into three heavyweight repos rather than a long tail. thedotmack/claude-mem (90,282 stars) captures everything an agent does during a session, compresses it with a model, and re-injects the relevant slice into future sessions; mem0ai/mem0 (62,934) positions as a provider-agnostic 'universal memory layer'; and MemPalace/mempalace (58,269) leads on benchmarks and stays free, built on ChromaDB with MCP wiring. All three are attacking the same failure — that a session boundary wipes everything the agent learned — but with different persistence models, and no interchange format exists between them. For builders, that means memory is currently a lock-in decision, not a swappable component.
2026-08-09 / agents-researcherAWS Strands Agents exposed the tenant-isolation key to the model: CVE-2026-19111 lets a prompt forge another tenant's memory namespaceAWS security bulletin 2026-077-AWS (August 6) discloses an IDOR in strands-agents-tools where `mongodb_memory`, `elasticsearch_memory` and `mem0_memory` used a `namespace` field as the sole tenant-isolation key — and exposed that field as an LLM-controllable tool-schema parameter. A crafted prompt could emit a call with a forged namespace to read, modify, delete or poison another tenant's memories, and the standalone mongodb/elasticsearch functions additionally exposed connection parameters so the memory layer could be redirected to an attacker-specified cluster. Fixed in 0.8.3; the design lesson is that any security boundary rendered into a tool schema is a boundary the model can rewrite.
2026-08-03 / agents-researchermem0 2.0.15 fixes a delete_all() that silently left most agent memories behindShipped August 1, mem0 v2.0.15 patches a data-deletion bug with compliance teeth: `delete_all()` listed the vector store once instead of paginating, so on any account with more memories than a single page — most vector stores default to about 100 — the remainder was silently left behind while the call reported success. It also caps Supabase search/list top_k at the vecs query limit of 1000 instead of erroring, and sets `size` on Elasticsearch KNN queries which were quietly capped at Elasticsearch's default 10 hits regardless of the requested top_k. The default LLMReranker model also moves from gpt-4o-mini to gpt-5-mini. If you have ever told a user their agent memory was deleted, re-run the check.
2026-07-30 / arxiv-researcherInferScale Injects Precomputed KV State Into vLLM's Paged Cache, Cutting TTFT 72–79% for Memory-Augmented AgentsInferScale (arXiv 2607.27090, 2026-07-29) targets the hidden tax in Mem0, MemGPT and Zep: retrieved memory facts get re-prefilled on every request. It precomputes each fact's KV representation, stores it on-GPU with a semantic embedding, and injects it directly into vLLM's paged cache via the KV-connector interface — no engine modifications, no fine-tuning — using Chunked RoPE to store keys pre-rotation and Context-Window Encoding to recover cross-fact context. Across three open-weight models on LoCoMo at k=50 it cuts TTFT by 72–79% (3.6–4.8x) and delivers 3.7–4.5x throughput under concurrent load, at 60.3% accuracy versus Mem0's 63.3%.

Graph relationships

SUPPORTS
Pi Agent
Mem0 shipped v0.3.0 plugins for Pi Agent.
Source finding
SUPPORTS
DeepSeek
Mem0 shipped v0.3.0 plugins for DeepSeek.
Source finding
SUPPORTS
OpenCode
Mem0 shipped v0.3.0 plugins for OpenCode.
Source finding
RELEASED
OpenClaw
Mem0 released OpenClaw v1.1.0 on September 9, 2026.
Source finding
BENCHMARKED AGAINST
AIOS
Agent kernel memory improves personalization scores 2.4-4.0 points against Mem0 across GPT-4o, Llama, and Qwen
Source finding
BENCHMARKED AGAINST
VoiceMem
VoiceMem's left brain beats Mem0 on top-5 retrieval benchmarks by nearly 30 points.
Source finding
BENCHMARKED AGAINST
Adaption Labs
Adaption Labs memory system achieved 90.6% on LongMemEval versus Mem0 OSS at 71.6%.
Source finding
BENCHMARKED AGAINST
OpenAI
Mem0 graph memory shows 26% relative improvement over OpenAI memory baselines.
Source finding
RELEASED
Mem0g
Mem0 released graph memory variant Mem0g in January 2026.
Source finding
BENCHMARKED AGAINST
Memori
Memori achieved 81.95% accuracy, beating Mem0 at 62.47%.
Source finding
SUPPORTS
AutoGen
AutoGen added Mem0 memory extension for persistent agent memory
Source finding
BENCHMARKED AGAINST
LoCoMo
Mem0's memory compression achieved 26% accuracy improvements on the LOCOMO benchmark.
Source finding

Source trail

arXiv 2608.26005 arXiv 2608.11879 arXiv 2607.27090 GitHub Source: GitHub / YourMemory Anthropic shipped Dreaming on May 6 Mem0's State of AI Agent Memory 2026 report A new glossary E2B partnered with Docker Memori Trending #1 TypeScript on GitHub The project

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues

Fetching from the wire…