docs: update README benchmarks to current temporal-reasoning results

Refresh the README benchmark table and Research Highlights to match the current numbers on the Memory Evaluation page (docs/core-concepts/ memory-evaluation.mdx), updated in #6056. - LoCoMo: 91.6 -> 92.5 - LongMemEval: 94.8 -> 94.4 (94.8 was not sourced by any docs page) - Assistant memory recall highlight: +53.6 delta -> 98.2 absolute - State the top_200 retrieval budget and the managed-platform caveat BEAM (1M / 10M), token counts, and latency are unchanged.
2026-07-09 15:50:16 +05:30
1 changed files with 5 additions and 5 deletions
@@ -46,12 +46,12 @@

 | Benchmark | Old | New  | Tokens  | Latency p50  |
 | --- | --- | --- | --- | --- |
-| **LoCoMo** | 71.4 | **91.6** | 7.0K  | 0.88s  |
-| **LongMemEval** | 67.8 | **94.8** | 6.8K  | 1.09s  |
+| **LoCoMo** | 71.4 | **92.5** | 7.0K  | 0.88s  |
+| **LongMemEval** | 67.8 | **94.4** | 6.8K  | 1.09s  |
 | **BEAM (1M)** | — | **64.1** | 6.7K  | 1.00s  |
 | **BEAM (10M)** | — | **48.6** | 6.9K  | 1.05s  |

-All benchmarks run on the same production-representative model stack. Single-pass retrieval (one call, no agentic loops).
+All benchmarks run on the same production-representative model stack. Single-pass retrieval (one call, no agentic loops) at a top_200 retrieval budget. Scores reflect Mem0's managed platform, which includes proprietary optimizations not available in the open-source SDK; open-source users should expect directionally similar gains but not identical numbers.

 **What changed:**
 - **Single-pass ADD-only extraction** -- one LLM call, no UPDATE/DELETE. Memories accumulate; nothing is overwritten.
@@ -63,8 +63,8 @@ All benchmarks run on the same production-representative model stack. Single-pas
 See the [migration guide](https://docs.mem0.ai/migration/oss-v2-to-v3) for upgrade instructions. The [evaluation framework](https://github.com/mem0ai/memory-benchmarks) is open-sourced so anyone can reproduce the numbers.

 ## Research Highlights
- **91.6 on LoCoMo** -- +20 points over the previous algorithm
- **94.8 on LongMemEval** -- +27 points, with +53.6 on assistant memory recall
+- **92.5 on LoCoMo** -- +21 points over the previous algorithm
+- **94.4 on LongMemEval** -- +27 points, with 98.2 on assistant memory recall
 - **64.1 on BEAM (1M)** -- production-scale memory evaluation at 1M tokens
 - [Read the full paper](https://mem0.ai/research)