← all runs
Cohere-10M — 2026-06-16
r8i.8xlarge · peak recall 0.9994 @ 3 QPS
QPS at recall
| Recall | Meerkat | vchordrq | vchordrq-rabitq8 | vchordrq-rabitq4 | Δ vs best competitor |
|---|
| 0.90 | 478 | 225 | 104 | 42 | +112% |
| 0.95 | 258 | 99 | 57 | 23 | +162% |
| 0.99 | 77 | 21 | 9 | 8 | +262% |
Latency at recall
| Meerkat | vchordrq | vchordrq-rabitq8 | vchordrq-rabitq4 | Δ vs best competitor |
|---|
| @ recall 0.90 | | | | | |
| p50 | 2.05 ms | 4.35 ms | 8.22 ms | 23.33 ms | -53% |
| p95 | 2.67 ms | 5.56 ms | 20.13 ms | 37.11 ms | -52% |
| p99 | 2.97 ms | 6.20 ms | 27.50 ms | 44.79 ms | -52% |
| @ recall 0.95 | | | | | |
| p50 | 3.80 ms | 9.94 ms | 14.27 ms | 42.49 ms | -62% |
| p95 | 4.81 ms | 12.51 ms | 37.59 ms | 70.40 ms | -62% |
| p99 | 5.39 ms | 13.85 ms | 51.61 ms | 84.79 ms | -61% |
| @ recall 0.99 | | | | | |
| p50 | 12.88 ms | 46.70 ms | 104.06 ms | 124.37 ms | -72% |
| p95 | 15.34 ms | 55.51 ms | 193.88 ms | 232.95 ms | -72% |
| p99 | 16.49 ms | 59.42 ms | 266.56 ms | 277.87 ms | -72% |
All values are taken at exactly the stated recall: each configuration's own sweep is log-interpolated between its two surrounding measured points (measured values are used verbatim on an exact hit) and the best configuration at that recall is shown — values are never interpolated across different configurations. “—” means no single sweep spans that recall; nothing is extrapolated. Recall is distance-based (ann-benchmarks, tie-aware). Insert/update/delete workloads are not measured.
Interpolated on config — Meerkat @0.90: mkt-n24000-cctrue-fs16-soar · vchordrq @0.90: vchordrq-n16000 · vchordrq-rabitq8 @0.90: vchordrq-rabitq8-n16000 · vchordrq-rabitq4 @0.90: vchordrq-rabitq4-rerank-n16000 · Meerkat @0.95: mkt-n24000-cctrue-fs16-soar · vchordrq @0.95: vchordrq-n16000 · vchordrq-rabitq8 @0.95: vchordrq-rabitq8-n16000 · vchordrq-rabitq4 @0.95: vchordrq-rabitq4-rerank-n16000 · Meerkat @0.99: mkt-n24000-cctrue-fs16-soar · vchordrq @0.99: vchordrq-n16000 · vchordrq-rabitq8 @0.99: vchordrq-rabitq8-n16000 · vchordrq-rabitq4 @0.99: vchordrq-rabitq4-rerank-n16000
Build comparison
| Meerkat | vchordrq | vchordrq-rabitq8 | vchordrq-rabitq4 | Δ vs best competitor |
|---|
| Build time | 632 s | 1216 s | 1544 s | 1744 s | -48% |
| Index size | 2.3 GB | 40.0 GB | 9.5 GB | 5.8 GB | -60% |
Build stats are for the configuration backing each column's best operating point: Meerkat = mkt-n24000-cctrue-fs16-soar · vchordrq = vchordrq-n16000 · vchordrq-rabitq8 = vchordrq-rabitq8-n16000 · vchordrq-rabitq4 = vchordrq-rabitq4-rerank-n16000.
Plots
Pareto svg ·
Distance-recall svg ·
Per-config svg ·
Build times svg ·
Index sizes svgHost
| Instance type | r8i.8xlarge |
| Hostname | ip-172-31-44-251 |
| CPU | Intel(R) Xeon(R) 6975P-C (32 vCPU) |
| Memory | 248 GB |
| OS | Ubuntu 26.04 LTS |
| Kernel | 7.0.0-1006-aws |
Versions
| rekall | 0.1.0 (e083fdb / main) |
| PostgreSQL | PostgreSQL 18.4 (Ubuntu 18.4-1.pgdg26.04+1) on x86_64-pc-linux-gnu, compiled by gcc (Ubuntu 15.2.0-16ubuntu1) 15.2.0, 64-bit |
| meerkat | 0.1.0 (cd8e895) |
| pageinspect | 1.13 |
| pg_prewarm | 1.2 |
| vchord | 1.1.1 |
| vector | 0.8.2 |
Search
Index builds
| Index | Type | Build time | Size | Workers |
|---|
| mkt-n24000-cctrue-fs16-soar | meerkat | 632.3s | 2.3 GB | 16 |
| vchordrq-n16000 | vchordrq | 1216.0s | 40.0 GB | 16 |
| vchordrq-rabitq4-rerank-n16000 | vchordrq-rabitq4 | 1743.9s | 5.8 GB | 16 |
| vchordrq-rabitq8-n16000 | vchordrq-rabitq8 | 1543.9s | 9.5 GB | 16 |
Index parameters
mkt-n24000-cctrue-fs16-soar: {"boundary_epsilon":0.05,"centroid_compression":true,"centroid_fastscan":true,"fan_out":null,"fastscan":true,"fastscan_bits":null,"nlist":24000,"rerank_vectors":null,"soar_lambda":1.0}
vchordrq-n16000: {"build_threads":16,"lists":[16000],"rerank_n":200,"residual_quantization":true,"spherical_centroids":true}
vchordrq-rabitq4-rerank-n16000: {"build_threads":16,"lists":[16000],"rerank_n":200,"residual_quantization":false,"spherical_centroids":true}
vchordrq-rabitq8-n16000: {"build_threads":16,"lists":[16000],"rerank_n":200,"residual_quantization":false,"spherical_centroids":true}
PostgreSQL settings
effective_cache_size = 192 GB
effective_io_concurrency = 256
huge_pages = try
maintenance_work_mem = 102 GB
max_connections = 100
max_parallel_maintenance_workers = 16
max_parallel_workers = 32
random_page_cost = 1.1
shared_buffers = 64 GB
wal_level = replica
work_mem = 1 GB
