vesper-memory 0.3.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/.env.example +28 -0
- package/CHANGELOG.md +137 -0
- package/LICENSE +21 -0
- package/README.md +665 -0
- package/config/sqlite-schema.sql +413 -0
- package/dist/benchmark/index.d.ts +18 -0
- package/dist/benchmark/index.d.ts.map +1 -0
- package/dist/benchmark/index.js +19 -0
- package/dist/benchmark/index.js.map +1 -0
- package/dist/benchmark/metrics-collector.d.ts +90 -0
- package/dist/benchmark/metrics-collector.d.ts.map +1 -0
- package/dist/benchmark/metrics-collector.js +196 -0
- package/dist/benchmark/metrics-collector.js.map +1 -0
- package/dist/benchmark/report-generator.d.ts +35 -0
- package/dist/benchmark/report-generator.d.ts.map +1 -0
- package/dist/benchmark/report-generator.js +221 -0
- package/dist/benchmark/report-generator.js.map +1 -0
- package/dist/benchmark/runner.d.ts +30 -0
- package/dist/benchmark/runner.d.ts.map +1 -0
- package/dist/benchmark/runner.js +308 -0
- package/dist/benchmark/runner.js.map +1 -0
- package/dist/benchmark/scenarios/context-recall.d.ts +65 -0
- package/dist/benchmark/scenarios/context-recall.d.ts.map +1 -0
- package/dist/benchmark/scenarios/context-recall.js +180 -0
- package/dist/benchmark/scenarios/context-recall.js.map +1 -0
- package/dist/benchmark/scenarios/index.d.ts +34 -0
- package/dist/benchmark/scenarios/index.d.ts.map +1 -0
- package/dist/benchmark/scenarios/index.js +230 -0
- package/dist/benchmark/scenarios/index.js.map +1 -0
- package/dist/benchmark/statistics.d.ts +125 -0
- package/dist/benchmark/statistics.d.ts.map +1 -0
- package/dist/benchmark/statistics.js +520 -0
- package/dist/benchmark/statistics.js.map +1 -0
- package/dist/benchmark/types.d.ts +458 -0
- package/dist/benchmark/types.d.ts.map +1 -0
- package/dist/benchmark/types.js +297 -0
- package/dist/benchmark/types.js.map +1 -0
- package/dist/cli 2.js +583 -0
- package/dist/cli.d 2.ts +8 -0
- package/dist/cli.d.ts +8 -0
- package/dist/cli.d.ts 2.map +1 -0
- package/dist/cli.d.ts.map +1 -0
- package/dist/cli.js +583 -0
- package/dist/cli.js 2.map +1 -0
- package/dist/cli.js.map +1 -0
- package/dist/config/rate-limits.d.ts +76 -0
- package/dist/config/rate-limits.d.ts.map +1 -0
- package/dist/config/rate-limits.js +161 -0
- package/dist/config/rate-limits.js.map +1 -0
- package/dist/config 2/rate-limits.d.ts +76 -0
- package/dist/config 2/rate-limits.d.ts.map +1 -0
- package/dist/config 2/rate-limits.js +161 -0
- package/dist/config 2/rate-limits.js.map +1 -0
- package/dist/consolidation/pipeline.d.ts +52 -0
- package/dist/consolidation/pipeline.d.ts.map +1 -0
- package/dist/consolidation/pipeline.js +193 -0
- package/dist/consolidation/pipeline.js.map +1 -0
- package/dist/consolidation 2/pipeline.d.ts +52 -0
- package/dist/consolidation 2/pipeline.d.ts.map +1 -0
- package/dist/consolidation 2/pipeline.js +193 -0
- package/dist/consolidation 2/pipeline.js.map +1 -0
- package/dist/embeddings/client.d.ts +95 -0
- package/dist/embeddings/client.d.ts.map +1 -0
- package/dist/embeddings/client.js +162 -0
- package/dist/embeddings/client.js.map +1 -0
- package/dist/embeddings 2/client.d.ts +94 -0
- package/dist/embeddings 2/client.d.ts.map +1 -0
- package/dist/embeddings 2/client.js +144 -0
- package/dist/embeddings 2/client.js.map +1 -0
- package/dist/memory-layers/semantic-memory.d.ts +76 -0
- package/dist/memory-layers/semantic-memory.d.ts.map +1 -0
- package/dist/memory-layers/semantic-memory.js +231 -0
- package/dist/memory-layers/semantic-memory.js.map +1 -0
- package/dist/memory-layers/skill-library.d.ts +25 -0
- package/dist/memory-layers/skill-library.d.ts.map +1 -0
- package/dist/memory-layers/skill-library.js +54 -0
- package/dist/memory-layers/skill-library.js.map +1 -0
- package/dist/memory-layers/working-memory.d.ts +57 -0
- package/dist/memory-layers/working-memory.d.ts.map +1 -0
- package/dist/memory-layers/working-memory.js +153 -0
- package/dist/memory-layers/working-memory.js.map +1 -0
- package/dist/memory-layers 2/semantic-memory.d.ts +33 -0
- package/dist/memory-layers 2/semantic-memory.d.ts.map +1 -0
- package/dist/memory-layers 2/semantic-memory.js +114 -0
- package/dist/memory-layers 2/semantic-memory.js.map +1 -0
- package/dist/memory-layers 2/skill-library.d.ts +25 -0
- package/dist/memory-layers 2/skill-library.d.ts.map +1 -0
- package/dist/memory-layers 2/skill-library.js +54 -0
- package/dist/memory-layers 2/skill-library.js.map +1 -0
- package/dist/memory-layers 2/working-memory.d.ts +57 -0
- package/dist/memory-layers 2/working-memory.d.ts.map +1 -0
- package/dist/memory-layers 2/working-memory.js +153 -0
- package/dist/memory-layers 2/working-memory.js.map +1 -0
- package/dist/middleware/auth.d.ts +287 -0
- package/dist/middleware/auth.d.ts.map +1 -0
- package/dist/middleware/auth.js +610 -0
- package/dist/middleware/auth.js.map +1 -0
- package/dist/middleware 2/auth.d.ts +287 -0
- package/dist/middleware 2/auth.d.ts.map +1 -0
- package/dist/middleware 2/auth.js +610 -0
- package/dist/middleware 2/auth.js.map +1 -0
- package/dist/monitoring/health.d.ts +241 -0
- package/dist/monitoring/health.d.ts.map +1 -0
- package/dist/monitoring/health.js +649 -0
- package/dist/monitoring/health.js.map +1 -0
- package/dist/monitoring/metrics.d.ts +153 -0
- package/dist/monitoring/metrics.d.ts.map +1 -0
- package/dist/monitoring/metrics.js +349 -0
- package/dist/monitoring/metrics.js.map +1 -0
- package/dist/monitoring 2/health.d.ts +241 -0
- package/dist/monitoring 2/health.d.ts.map +1 -0
- package/dist/monitoring 2/health.js +649 -0
- package/dist/monitoring 2/health.js.map +1 -0
- package/dist/monitoring 2/metrics.d.ts +153 -0
- package/dist/monitoring 2/metrics.d.ts.map +1 -0
- package/dist/monitoring 2/metrics.js +349 -0
- package/dist/monitoring 2/metrics.js.map +1 -0
- package/dist/retrieval/hybrid-search.d.ts +186 -0
- package/dist/retrieval/hybrid-search.d.ts.map +1 -0
- package/dist/retrieval/hybrid-search.js +330 -0
- package/dist/retrieval/hybrid-search.js.map +1 -0
- package/dist/retrieval 2/hybrid-search.d.ts +186 -0
- package/dist/retrieval 2/hybrid-search.d.ts.map +1 -0
- package/dist/retrieval 2/hybrid-search.js +327 -0
- package/dist/retrieval 2/hybrid-search.js.map +1 -0
- package/dist/router/smart-router.d.ts +157 -0
- package/dist/router/smart-router.d.ts.map +1 -0
- package/dist/router/smart-router.js +524 -0
- package/dist/router/smart-router.js.map +1 -0
- package/dist/router 2/smart-router.d.ts +111 -0
- package/dist/router 2/smart-router.d.ts.map +1 -0
- package/dist/router 2/smart-router.js +383 -0
- package/dist/router 2/smart-router.js.map +1 -0
- package/dist/security/rate-limit-middleware.d.ts +119 -0
- package/dist/security/rate-limit-middleware.d.ts.map +1 -0
- package/dist/security/rate-limit-middleware.js +193 -0
- package/dist/security/rate-limit-middleware.js.map +1 -0
- package/dist/security 2/rate-limit-middleware.d.ts +119 -0
- package/dist/security 2/rate-limit-middleware.d.ts.map +1 -0
- package/dist/security 2/rate-limit-middleware.js +193 -0
- package/dist/security 2/rate-limit-middleware.js.map +1 -0
- package/dist/server 2.js +861 -0
- package/dist/server-wrapper 2.js +170 -0
- package/dist/server-wrapper.d 2.ts +12 -0
- package/dist/server-wrapper.d.ts +12 -0
- package/dist/server-wrapper.d.ts 2.map +1 -0
- package/dist/server-wrapper.d.ts.map +1 -0
- package/dist/server-wrapper.js +170 -0
- package/dist/server-wrapper.js 2.map +1 -0
- package/dist/server-wrapper.js.map +1 -0
- package/dist/server.d 2.ts +14 -0
- package/dist/server.d.ts +14 -0
- package/dist/server.d.ts 2.map +1 -0
- package/dist/server.d.ts.map +1 -0
- package/dist/server.js +656 -0
- package/dist/server.js 2.map +1 -0
- package/dist/server.js.map +1 -0
- package/dist/synthesis/conflict-detector.d.ts +43 -0
- package/dist/synthesis/conflict-detector.d.ts.map +1 -0
- package/dist/synthesis/conflict-detector.js +125 -0
- package/dist/synthesis/conflict-detector.js.map +1 -0
- package/dist/synthesis 2/conflict-detector.d.ts +43 -0
- package/dist/synthesis 2/conflict-detector.d.ts.map +1 -0
- package/dist/synthesis 2/conflict-detector.js +125 -0
- package/dist/synthesis 2/conflict-detector.js.map +1 -0
- package/dist/utils/monitoring.d.ts +25 -0
- package/dist/utils/monitoring.d.ts.map +1 -0
- package/dist/utils/monitoring.js +41 -0
- package/dist/utils/monitoring.js.map +1 -0
- package/dist/utils/rate-limiter.d.ts +97 -0
- package/dist/utils/rate-limiter.d.ts.map +1 -0
- package/dist/utils/rate-limiter.js +166 -0
- package/dist/utils/rate-limiter.js.map +1 -0
- package/dist/utils/validation.d.ts +139 -0
- package/dist/utils/validation.d.ts.map +1 -0
- package/dist/utils/validation.js +195 -0
- package/dist/utils/validation.js.map +1 -0
- package/dist/utils 2/monitoring.d.ts +25 -0
- package/dist/utils 2/monitoring.d.ts.map +1 -0
- package/dist/utils 2/monitoring.js +41 -0
- package/dist/utils 2/monitoring.js.map +1 -0
- package/dist/utils 2/rate-limiter.d.ts +97 -0
- package/dist/utils 2/rate-limiter.d.ts.map +1 -0
- package/dist/utils 2/rate-limiter.js +166 -0
- package/dist/utils 2/rate-limiter.js.map +1 -0
- package/dist/utils 2/validation.d.ts +139 -0
- package/dist/utils 2/validation.d.ts.map +1 -0
- package/dist/utils 2/validation.js +195 -0
- package/dist/utils 2/validation.js.map +1 -0
- package/docker-compose.yml +65 -0
- package/embedding-service/Dockerfile +29 -0
- package/embedding-service/README.md +68 -0
- package/embedding-service/requirements.txt +6 -0
- package/embedding-service/server.py +110 -0
- package/package.json +92 -0
- package/scripts/FRESH_INSTALL.md +280 -0
- package/scripts/README.md +101 -0
- package/scripts/ensure-infrastructure.sh +38 -0
- package/scripts/fresh-install-auto.sh +96 -0
- package/scripts/fresh-install.sh +167 -0
- package/scripts/health-check.sh +72 -0
- package/scripts/init-qdrant.ts +201 -0
- package/scripts/postinstall.js +94 -0
- package/scripts/run-accuracy-benchmark.ts +1237 -0
- package/scripts/run-real-benchmark.ts +581 -0
- package/scripts/test-mcp-server.sh +38 -0
|
@@ -0,0 +1,458 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* Type Definitions for Scientific Benchmarks
|
|
3
|
+
*
|
|
4
|
+
* Comprehensive type system with runtime validation using Zod.
|
|
5
|
+
* Ensures type safety at compile time and runtime.
|
|
6
|
+
*/
|
|
7
|
+
import { z } from "zod";
|
|
8
|
+
/**
|
|
9
|
+
* Available benchmark scenario types
|
|
10
|
+
*/
|
|
11
|
+
export declare const ScenarioTypeEnum: z.ZodEnum<{
|
|
12
|
+
"context-recall": "context-recall";
|
|
13
|
+
"cross-session": "cross-session";
|
|
14
|
+
"skill-retrieval": "skill-retrieval";
|
|
15
|
+
"token-efficiency": "token-efficiency";
|
|
16
|
+
"semantic-accuracy": "semantic-accuracy";
|
|
17
|
+
}>;
|
|
18
|
+
export type ScenarioType = z.infer<typeof ScenarioTypeEnum>;
|
|
19
|
+
/**
|
|
20
|
+
* Benchmark phases for A/B comparison
|
|
21
|
+
*/
|
|
22
|
+
export declare const BenchmarkPhaseEnum: z.ZodEnum<{
|
|
23
|
+
"vesper-enabled": "vesper-enabled";
|
|
24
|
+
"vesper-disabled": "vesper-disabled";
|
|
25
|
+
}>;
|
|
26
|
+
export type BenchmarkPhase = z.infer<typeof BenchmarkPhaseEnum>;
|
|
27
|
+
/**
|
|
28
|
+
* Winner determination
|
|
29
|
+
*/
|
|
30
|
+
export declare const WinnerEnum: z.ZodEnum<{
|
|
31
|
+
"vesper-enabled": "vesper-enabled";
|
|
32
|
+
"vesper-disabled": "vesper-disabled";
|
|
33
|
+
tie: "tie";
|
|
34
|
+
}>;
|
|
35
|
+
export type Winner = z.infer<typeof WinnerEnum>;
|
|
36
|
+
/**
|
|
37
|
+
* Configuration for running benchmarks
|
|
38
|
+
*/
|
|
39
|
+
export declare const BenchmarkConfigSchema: z.ZodObject<{
|
|
40
|
+
warmupRuns: z.ZodDefault<z.ZodNumber>;
|
|
41
|
+
measurementRuns: z.ZodDefault<z.ZodNumber>;
|
|
42
|
+
timeoutMs: z.ZodDefault<z.ZodNumber>;
|
|
43
|
+
scenarios: z.ZodArray<z.ZodEnum<{
|
|
44
|
+
"context-recall": "context-recall";
|
|
45
|
+
"cross-session": "cross-session";
|
|
46
|
+
"skill-retrieval": "skill-retrieval";
|
|
47
|
+
"token-efficiency": "token-efficiency";
|
|
48
|
+
"semantic-accuracy": "semantic-accuracy";
|
|
49
|
+
}>>;
|
|
50
|
+
outputDirectory: z.ZodDefault<z.ZodString>;
|
|
51
|
+
significanceLevel: z.ZodDefault<z.ZodNumber>;
|
|
52
|
+
}, z.core.$strip>;
|
|
53
|
+
export type BenchmarkConfig = z.infer<typeof BenchmarkConfigSchema>;
|
|
54
|
+
/**
|
|
55
|
+
* A single metrics measurement point
|
|
56
|
+
*/
|
|
57
|
+
export declare const MetricsSnapshotSchema: z.ZodObject<{
|
|
58
|
+
timestamp: z.ZodNumber;
|
|
59
|
+
latencyMs: z.ZodNumber;
|
|
60
|
+
tokensUsed: z.ZodOptional<z.ZodNumber>;
|
|
61
|
+
retrievalAccuracy: z.ZodOptional<z.ZodNumber>;
|
|
62
|
+
memoryHit: z.ZodBoolean;
|
|
63
|
+
responseQuality: z.ZodOptional<z.ZodNumber>;
|
|
64
|
+
metadata: z.ZodOptional<z.ZodRecord<z.ZodString, z.ZodUnknown>>;
|
|
65
|
+
}, z.core.$strip>;
|
|
66
|
+
export type MetricsSnapshot = z.infer<typeof MetricsSnapshotSchema>;
|
|
67
|
+
/**
|
|
68
|
+
* Aggregated metrics from multiple runs
|
|
69
|
+
*/
|
|
70
|
+
export declare const AggregateMetricsSchema: z.ZodObject<{
|
|
71
|
+
latencyP50: z.ZodNumber;
|
|
72
|
+
latencyP95: z.ZodNumber;
|
|
73
|
+
latencyP99: z.ZodNumber;
|
|
74
|
+
avgTokens: z.ZodOptional<z.ZodNumber>;
|
|
75
|
+
avgAccuracy: z.ZodOptional<z.ZodNumber>;
|
|
76
|
+
memoryHitRate: z.ZodOptional<z.ZodNumber>;
|
|
77
|
+
latencyMean: z.ZodOptional<z.ZodNumber>;
|
|
78
|
+
latencyStdDev: z.ZodOptional<z.ZodNumber>;
|
|
79
|
+
}, z.core.$strip>;
|
|
80
|
+
export type AggregateMetrics = z.infer<typeof AggregateMetricsSchema>;
|
|
81
|
+
/**
|
|
82
|
+
* Result from running a single scenario in one phase
|
|
83
|
+
*/
|
|
84
|
+
export declare const ScenarioResultSchema: z.ZodObject<{
|
|
85
|
+
scenarioType: z.ZodEnum<{
|
|
86
|
+
"context-recall": "context-recall";
|
|
87
|
+
"cross-session": "cross-session";
|
|
88
|
+
"skill-retrieval": "skill-retrieval";
|
|
89
|
+
"token-efficiency": "token-efficiency";
|
|
90
|
+
"semantic-accuracy": "semantic-accuracy";
|
|
91
|
+
}>;
|
|
92
|
+
phase: z.ZodEnum<{
|
|
93
|
+
"vesper-enabled": "vesper-enabled";
|
|
94
|
+
"vesper-disabled": "vesper-disabled";
|
|
95
|
+
}>;
|
|
96
|
+
metrics: z.ZodArray<z.ZodObject<{
|
|
97
|
+
timestamp: z.ZodNumber;
|
|
98
|
+
latencyMs: z.ZodNumber;
|
|
99
|
+
tokensUsed: z.ZodOptional<z.ZodNumber>;
|
|
100
|
+
retrievalAccuracy: z.ZodOptional<z.ZodNumber>;
|
|
101
|
+
memoryHit: z.ZodBoolean;
|
|
102
|
+
responseQuality: z.ZodOptional<z.ZodNumber>;
|
|
103
|
+
metadata: z.ZodOptional<z.ZodRecord<z.ZodString, z.ZodUnknown>>;
|
|
104
|
+
}, z.core.$strip>>;
|
|
105
|
+
aggregates: z.ZodObject<{
|
|
106
|
+
latencyP50: z.ZodNumber;
|
|
107
|
+
latencyP95: z.ZodNumber;
|
|
108
|
+
latencyP99: z.ZodNumber;
|
|
109
|
+
avgTokens: z.ZodOptional<z.ZodNumber>;
|
|
110
|
+
avgAccuracy: z.ZodOptional<z.ZodNumber>;
|
|
111
|
+
memoryHitRate: z.ZodOptional<z.ZodNumber>;
|
|
112
|
+
latencyMean: z.ZodOptional<z.ZodNumber>;
|
|
113
|
+
latencyStdDev: z.ZodOptional<z.ZodNumber>;
|
|
114
|
+
}, z.core.$strip>;
|
|
115
|
+
rawResponses: z.ZodArray<z.ZodString>;
|
|
116
|
+
errors: z.ZodOptional<z.ZodArray<z.ZodString>>;
|
|
117
|
+
}, z.core.$strip>;
|
|
118
|
+
export type ScenarioResult = z.infer<typeof ScenarioResultSchema>;
|
|
119
|
+
/**
|
|
120
|
+
* T-test result structure
|
|
121
|
+
*/
|
|
122
|
+
export declare const TTestResultSchema: z.ZodObject<{
|
|
123
|
+
tStatistic: z.ZodNumber;
|
|
124
|
+
pValue: z.ZodNumber;
|
|
125
|
+
degreesOfFreedom: z.ZodNumber;
|
|
126
|
+
significant: z.ZodBoolean;
|
|
127
|
+
alpha: z.ZodNumber;
|
|
128
|
+
meanDifference: z.ZodNumber;
|
|
129
|
+
}, z.core.$strip>;
|
|
130
|
+
export type TTestResult = z.infer<typeof TTestResultSchema>;
|
|
131
|
+
/**
|
|
132
|
+
* Effect size result
|
|
133
|
+
*/
|
|
134
|
+
export declare const EffectSizeSchema: z.ZodObject<{
|
|
135
|
+
cohensD: z.ZodNumber;
|
|
136
|
+
interpretation: z.ZodEnum<{
|
|
137
|
+
small: "small";
|
|
138
|
+
negligible: "negligible";
|
|
139
|
+
medium: "medium";
|
|
140
|
+
large: "large";
|
|
141
|
+
}>;
|
|
142
|
+
}, z.core.$strip>;
|
|
143
|
+
export type EffectSize = z.infer<typeof EffectSizeSchema>;
|
|
144
|
+
/**
|
|
145
|
+
* Statistical comparison between enabled and disabled
|
|
146
|
+
*/
|
|
147
|
+
export declare const StatisticalComparisonSchema: z.ZodObject<{
|
|
148
|
+
latencyImprovement: z.ZodNumber;
|
|
149
|
+
tokenSavings: z.ZodNumber;
|
|
150
|
+
accuracyDelta: z.ZodNumber;
|
|
151
|
+
tTestResult: z.ZodObject<{
|
|
152
|
+
tStatistic: z.ZodNumber;
|
|
153
|
+
pValue: z.ZodNumber;
|
|
154
|
+
degreesOfFreedom: z.ZodNumber;
|
|
155
|
+
significant: z.ZodBoolean;
|
|
156
|
+
alpha: z.ZodNumber;
|
|
157
|
+
meanDifference: z.ZodNumber;
|
|
158
|
+
}, z.core.$strip>;
|
|
159
|
+
effectSize: z.ZodObject<{
|
|
160
|
+
cohensD: z.ZodNumber;
|
|
161
|
+
interpretation: z.ZodEnum<{
|
|
162
|
+
small: "small";
|
|
163
|
+
negligible: "negligible";
|
|
164
|
+
medium: "medium";
|
|
165
|
+
large: "large";
|
|
166
|
+
}>;
|
|
167
|
+
}, z.core.$strip>;
|
|
168
|
+
}, z.core.$strip>;
|
|
169
|
+
export type StatisticalComparison = z.infer<typeof StatisticalComparisonSchema>;
|
|
170
|
+
/**
|
|
171
|
+
* Comparison between Vesper-enabled and Vesper-disabled for a scenario
|
|
172
|
+
*/
|
|
173
|
+
export declare const ComparisonResultSchema: z.ZodObject<{
|
|
174
|
+
scenarioType: z.ZodEnum<{
|
|
175
|
+
"context-recall": "context-recall";
|
|
176
|
+
"cross-session": "cross-session";
|
|
177
|
+
"skill-retrieval": "skill-retrieval";
|
|
178
|
+
"token-efficiency": "token-efficiency";
|
|
179
|
+
"semantic-accuracy": "semantic-accuracy";
|
|
180
|
+
}>;
|
|
181
|
+
enabledResult: z.ZodObject<{
|
|
182
|
+
scenarioType: z.ZodEnum<{
|
|
183
|
+
"context-recall": "context-recall";
|
|
184
|
+
"cross-session": "cross-session";
|
|
185
|
+
"skill-retrieval": "skill-retrieval";
|
|
186
|
+
"token-efficiency": "token-efficiency";
|
|
187
|
+
"semantic-accuracy": "semantic-accuracy";
|
|
188
|
+
}>;
|
|
189
|
+
phase: z.ZodEnum<{
|
|
190
|
+
"vesper-enabled": "vesper-enabled";
|
|
191
|
+
"vesper-disabled": "vesper-disabled";
|
|
192
|
+
}>;
|
|
193
|
+
metrics: z.ZodArray<z.ZodObject<{
|
|
194
|
+
timestamp: z.ZodNumber;
|
|
195
|
+
latencyMs: z.ZodNumber;
|
|
196
|
+
tokensUsed: z.ZodOptional<z.ZodNumber>;
|
|
197
|
+
retrievalAccuracy: z.ZodOptional<z.ZodNumber>;
|
|
198
|
+
memoryHit: z.ZodBoolean;
|
|
199
|
+
responseQuality: z.ZodOptional<z.ZodNumber>;
|
|
200
|
+
metadata: z.ZodOptional<z.ZodRecord<z.ZodString, z.ZodUnknown>>;
|
|
201
|
+
}, z.core.$strip>>;
|
|
202
|
+
aggregates: z.ZodObject<{
|
|
203
|
+
latencyP50: z.ZodNumber;
|
|
204
|
+
latencyP95: z.ZodNumber;
|
|
205
|
+
latencyP99: z.ZodNumber;
|
|
206
|
+
avgTokens: z.ZodOptional<z.ZodNumber>;
|
|
207
|
+
avgAccuracy: z.ZodOptional<z.ZodNumber>;
|
|
208
|
+
memoryHitRate: z.ZodOptional<z.ZodNumber>;
|
|
209
|
+
latencyMean: z.ZodOptional<z.ZodNumber>;
|
|
210
|
+
latencyStdDev: z.ZodOptional<z.ZodNumber>;
|
|
211
|
+
}, z.core.$strip>;
|
|
212
|
+
rawResponses: z.ZodArray<z.ZodString>;
|
|
213
|
+
errors: z.ZodOptional<z.ZodArray<z.ZodString>>;
|
|
214
|
+
}, z.core.$strip>;
|
|
215
|
+
disabledResult: z.ZodObject<{
|
|
216
|
+
scenarioType: z.ZodEnum<{
|
|
217
|
+
"context-recall": "context-recall";
|
|
218
|
+
"cross-session": "cross-session";
|
|
219
|
+
"skill-retrieval": "skill-retrieval";
|
|
220
|
+
"token-efficiency": "token-efficiency";
|
|
221
|
+
"semantic-accuracy": "semantic-accuracy";
|
|
222
|
+
}>;
|
|
223
|
+
phase: z.ZodEnum<{
|
|
224
|
+
"vesper-enabled": "vesper-enabled";
|
|
225
|
+
"vesper-disabled": "vesper-disabled";
|
|
226
|
+
}>;
|
|
227
|
+
metrics: z.ZodArray<z.ZodObject<{
|
|
228
|
+
timestamp: z.ZodNumber;
|
|
229
|
+
latencyMs: z.ZodNumber;
|
|
230
|
+
tokensUsed: z.ZodOptional<z.ZodNumber>;
|
|
231
|
+
retrievalAccuracy: z.ZodOptional<z.ZodNumber>;
|
|
232
|
+
memoryHit: z.ZodBoolean;
|
|
233
|
+
responseQuality: z.ZodOptional<z.ZodNumber>;
|
|
234
|
+
metadata: z.ZodOptional<z.ZodRecord<z.ZodString, z.ZodUnknown>>;
|
|
235
|
+
}, z.core.$strip>>;
|
|
236
|
+
aggregates: z.ZodObject<{
|
|
237
|
+
latencyP50: z.ZodNumber;
|
|
238
|
+
latencyP95: z.ZodNumber;
|
|
239
|
+
latencyP99: z.ZodNumber;
|
|
240
|
+
avgTokens: z.ZodOptional<z.ZodNumber>;
|
|
241
|
+
avgAccuracy: z.ZodOptional<z.ZodNumber>;
|
|
242
|
+
memoryHitRate: z.ZodOptional<z.ZodNumber>;
|
|
243
|
+
latencyMean: z.ZodOptional<z.ZodNumber>;
|
|
244
|
+
latencyStdDev: z.ZodOptional<z.ZodNumber>;
|
|
245
|
+
}, z.core.$strip>;
|
|
246
|
+
rawResponses: z.ZodArray<z.ZodString>;
|
|
247
|
+
errors: z.ZodOptional<z.ZodArray<z.ZodString>>;
|
|
248
|
+
}, z.core.$strip>;
|
|
249
|
+
statistics: z.ZodObject<{
|
|
250
|
+
latencyImprovement: z.ZodNumber;
|
|
251
|
+
tokenSavings: z.ZodNumber;
|
|
252
|
+
accuracyDelta: z.ZodNumber;
|
|
253
|
+
tTestResult: z.ZodObject<{
|
|
254
|
+
tStatistic: z.ZodNumber;
|
|
255
|
+
pValue: z.ZodNumber;
|
|
256
|
+
degreesOfFreedom: z.ZodNumber;
|
|
257
|
+
significant: z.ZodBoolean;
|
|
258
|
+
alpha: z.ZodNumber;
|
|
259
|
+
meanDifference: z.ZodNumber;
|
|
260
|
+
}, z.core.$strip>;
|
|
261
|
+
effectSize: z.ZodObject<{
|
|
262
|
+
cohensD: z.ZodNumber;
|
|
263
|
+
interpretation: z.ZodEnum<{
|
|
264
|
+
small: "small";
|
|
265
|
+
negligible: "negligible";
|
|
266
|
+
medium: "medium";
|
|
267
|
+
large: "large";
|
|
268
|
+
}>;
|
|
269
|
+
}, z.core.$strip>;
|
|
270
|
+
}, z.core.$strip>;
|
|
271
|
+
winner: z.ZodEnum<{
|
|
272
|
+
"vesper-enabled": "vesper-enabled";
|
|
273
|
+
"vesper-disabled": "vesper-disabled";
|
|
274
|
+
tie: "tie";
|
|
275
|
+
}>;
|
|
276
|
+
}, z.core.$strip>;
|
|
277
|
+
export type ComparisonResult = z.infer<typeof ComparisonResultSchema>;
|
|
278
|
+
/**
|
|
279
|
+
* Summary of all benchmark results
|
|
280
|
+
*/
|
|
281
|
+
export declare const BenchmarkSummarySchema: z.ZodObject<{
|
|
282
|
+
totalScenarios: z.ZodNumber;
|
|
283
|
+
vesperWins: z.ZodNumber;
|
|
284
|
+
disabledWins: z.ZodNumber;
|
|
285
|
+
ties: z.ZodNumber;
|
|
286
|
+
overallImprovement: z.ZodNumber;
|
|
287
|
+
statisticallySignificant: z.ZodBoolean;
|
|
288
|
+
}, z.core.$strip>;
|
|
289
|
+
export type BenchmarkSummary = z.infer<typeof BenchmarkSummarySchema>;
|
|
290
|
+
/**
|
|
291
|
+
* Complete benchmark result with all data
|
|
292
|
+
*/
|
|
293
|
+
export declare const BenchmarkResultSchema: z.ZodObject<{
|
|
294
|
+
id: z.ZodString;
|
|
295
|
+
startTime: z.ZodNumber;
|
|
296
|
+
endTime: z.ZodNumber;
|
|
297
|
+
config: z.ZodObject<{
|
|
298
|
+
warmupRuns: z.ZodDefault<z.ZodNumber>;
|
|
299
|
+
measurementRuns: z.ZodDefault<z.ZodNumber>;
|
|
300
|
+
timeoutMs: z.ZodDefault<z.ZodNumber>;
|
|
301
|
+
scenarios: z.ZodArray<z.ZodEnum<{
|
|
302
|
+
"context-recall": "context-recall";
|
|
303
|
+
"cross-session": "cross-session";
|
|
304
|
+
"skill-retrieval": "skill-retrieval";
|
|
305
|
+
"token-efficiency": "token-efficiency";
|
|
306
|
+
"semantic-accuracy": "semantic-accuracy";
|
|
307
|
+
}>>;
|
|
308
|
+
outputDirectory: z.ZodDefault<z.ZodString>;
|
|
309
|
+
significanceLevel: z.ZodDefault<z.ZodNumber>;
|
|
310
|
+
}, z.core.$strip>;
|
|
311
|
+
comparisons: z.ZodArray<z.ZodObject<{
|
|
312
|
+
scenarioType: z.ZodEnum<{
|
|
313
|
+
"context-recall": "context-recall";
|
|
314
|
+
"cross-session": "cross-session";
|
|
315
|
+
"skill-retrieval": "skill-retrieval";
|
|
316
|
+
"token-efficiency": "token-efficiency";
|
|
317
|
+
"semantic-accuracy": "semantic-accuracy";
|
|
318
|
+
}>;
|
|
319
|
+
enabledResult: z.ZodObject<{
|
|
320
|
+
scenarioType: z.ZodEnum<{
|
|
321
|
+
"context-recall": "context-recall";
|
|
322
|
+
"cross-session": "cross-session";
|
|
323
|
+
"skill-retrieval": "skill-retrieval";
|
|
324
|
+
"token-efficiency": "token-efficiency";
|
|
325
|
+
"semantic-accuracy": "semantic-accuracy";
|
|
326
|
+
}>;
|
|
327
|
+
phase: z.ZodEnum<{
|
|
328
|
+
"vesper-enabled": "vesper-enabled";
|
|
329
|
+
"vesper-disabled": "vesper-disabled";
|
|
330
|
+
}>;
|
|
331
|
+
metrics: z.ZodArray<z.ZodObject<{
|
|
332
|
+
timestamp: z.ZodNumber;
|
|
333
|
+
latencyMs: z.ZodNumber;
|
|
334
|
+
tokensUsed: z.ZodOptional<z.ZodNumber>;
|
|
335
|
+
retrievalAccuracy: z.ZodOptional<z.ZodNumber>;
|
|
336
|
+
memoryHit: z.ZodBoolean;
|
|
337
|
+
responseQuality: z.ZodOptional<z.ZodNumber>;
|
|
338
|
+
metadata: z.ZodOptional<z.ZodRecord<z.ZodString, z.ZodUnknown>>;
|
|
339
|
+
}, z.core.$strip>>;
|
|
340
|
+
aggregates: z.ZodObject<{
|
|
341
|
+
latencyP50: z.ZodNumber;
|
|
342
|
+
latencyP95: z.ZodNumber;
|
|
343
|
+
latencyP99: z.ZodNumber;
|
|
344
|
+
avgTokens: z.ZodOptional<z.ZodNumber>;
|
|
345
|
+
avgAccuracy: z.ZodOptional<z.ZodNumber>;
|
|
346
|
+
memoryHitRate: z.ZodOptional<z.ZodNumber>;
|
|
347
|
+
latencyMean: z.ZodOptional<z.ZodNumber>;
|
|
348
|
+
latencyStdDev: z.ZodOptional<z.ZodNumber>;
|
|
349
|
+
}, z.core.$strip>;
|
|
350
|
+
rawResponses: z.ZodArray<z.ZodString>;
|
|
351
|
+
errors: z.ZodOptional<z.ZodArray<z.ZodString>>;
|
|
352
|
+
}, z.core.$strip>;
|
|
353
|
+
disabledResult: z.ZodObject<{
|
|
354
|
+
scenarioType: z.ZodEnum<{
|
|
355
|
+
"context-recall": "context-recall";
|
|
356
|
+
"cross-session": "cross-session";
|
|
357
|
+
"skill-retrieval": "skill-retrieval";
|
|
358
|
+
"token-efficiency": "token-efficiency";
|
|
359
|
+
"semantic-accuracy": "semantic-accuracy";
|
|
360
|
+
}>;
|
|
361
|
+
phase: z.ZodEnum<{
|
|
362
|
+
"vesper-enabled": "vesper-enabled";
|
|
363
|
+
"vesper-disabled": "vesper-disabled";
|
|
364
|
+
}>;
|
|
365
|
+
metrics: z.ZodArray<z.ZodObject<{
|
|
366
|
+
timestamp: z.ZodNumber;
|
|
367
|
+
latencyMs: z.ZodNumber;
|
|
368
|
+
tokensUsed: z.ZodOptional<z.ZodNumber>;
|
|
369
|
+
retrievalAccuracy: z.ZodOptional<z.ZodNumber>;
|
|
370
|
+
memoryHit: z.ZodBoolean;
|
|
371
|
+
responseQuality: z.ZodOptional<z.ZodNumber>;
|
|
372
|
+
metadata: z.ZodOptional<z.ZodRecord<z.ZodString, z.ZodUnknown>>;
|
|
373
|
+
}, z.core.$strip>>;
|
|
374
|
+
aggregates: z.ZodObject<{
|
|
375
|
+
latencyP50: z.ZodNumber;
|
|
376
|
+
latencyP95: z.ZodNumber;
|
|
377
|
+
latencyP99: z.ZodNumber;
|
|
378
|
+
avgTokens: z.ZodOptional<z.ZodNumber>;
|
|
379
|
+
avgAccuracy: z.ZodOptional<z.ZodNumber>;
|
|
380
|
+
memoryHitRate: z.ZodOptional<z.ZodNumber>;
|
|
381
|
+
latencyMean: z.ZodOptional<z.ZodNumber>;
|
|
382
|
+
latencyStdDev: z.ZodOptional<z.ZodNumber>;
|
|
383
|
+
}, z.core.$strip>;
|
|
384
|
+
rawResponses: z.ZodArray<z.ZodString>;
|
|
385
|
+
errors: z.ZodOptional<z.ZodArray<z.ZodString>>;
|
|
386
|
+
}, z.core.$strip>;
|
|
387
|
+
statistics: z.ZodObject<{
|
|
388
|
+
latencyImprovement: z.ZodNumber;
|
|
389
|
+
tokenSavings: z.ZodNumber;
|
|
390
|
+
accuracyDelta: z.ZodNumber;
|
|
391
|
+
tTestResult: z.ZodObject<{
|
|
392
|
+
tStatistic: z.ZodNumber;
|
|
393
|
+
pValue: z.ZodNumber;
|
|
394
|
+
degreesOfFreedom: z.ZodNumber;
|
|
395
|
+
significant: z.ZodBoolean;
|
|
396
|
+
alpha: z.ZodNumber;
|
|
397
|
+
meanDifference: z.ZodNumber;
|
|
398
|
+
}, z.core.$strip>;
|
|
399
|
+
effectSize: z.ZodObject<{
|
|
400
|
+
cohensD: z.ZodNumber;
|
|
401
|
+
interpretation: z.ZodEnum<{
|
|
402
|
+
small: "small";
|
|
403
|
+
negligible: "negligible";
|
|
404
|
+
medium: "medium";
|
|
405
|
+
large: "large";
|
|
406
|
+
}>;
|
|
407
|
+
}, z.core.$strip>;
|
|
408
|
+
}, z.core.$strip>;
|
|
409
|
+
winner: z.ZodEnum<{
|
|
410
|
+
"vesper-enabled": "vesper-enabled";
|
|
411
|
+
"vesper-disabled": "vesper-disabled";
|
|
412
|
+
tie: "tie";
|
|
413
|
+
}>;
|
|
414
|
+
}, z.core.$strip>>;
|
|
415
|
+
summary: z.ZodObject<{
|
|
416
|
+
totalScenarios: z.ZodNumber;
|
|
417
|
+
vesperWins: z.ZodNumber;
|
|
418
|
+
disabledWins: z.ZodNumber;
|
|
419
|
+
ties: z.ZodNumber;
|
|
420
|
+
overallImprovement: z.ZodNumber;
|
|
421
|
+
statisticallySignificant: z.ZodBoolean;
|
|
422
|
+
}, z.core.$strip>;
|
|
423
|
+
metadata: z.ZodObject<{
|
|
424
|
+
version: z.ZodString;
|
|
425
|
+
environment: z.ZodString;
|
|
426
|
+
hostname: z.ZodOptional<z.ZodString>;
|
|
427
|
+
nodeVersion: z.ZodOptional<z.ZodString>;
|
|
428
|
+
}, z.core.$strip>;
|
|
429
|
+
}, z.core.$strip>;
|
|
430
|
+
export type BenchmarkResult = z.infer<typeof BenchmarkResultSchema>;
|
|
431
|
+
/**
|
|
432
|
+
* Create a default benchmark configuration
|
|
433
|
+
*/
|
|
434
|
+
export declare function createDefaultConfig(overrides?: Partial<BenchmarkConfig>): BenchmarkConfig;
|
|
435
|
+
/**
|
|
436
|
+
* Validate and parse a benchmark configuration
|
|
437
|
+
* @throws ZodError if validation fails
|
|
438
|
+
*/
|
|
439
|
+
export declare function validateBenchmarkConfig(input: unknown): BenchmarkConfig;
|
|
440
|
+
/**
|
|
441
|
+
* Validate and parse a benchmark result
|
|
442
|
+
* @throws ZodError if validation fails
|
|
443
|
+
*/
|
|
444
|
+
export declare function validateBenchmarkResult(input: unknown): BenchmarkResult;
|
|
445
|
+
/**
|
|
446
|
+
* Validate and parse a metrics snapshot
|
|
447
|
+
* @throws ZodError if validation fails
|
|
448
|
+
*/
|
|
449
|
+
export declare function validateMetricsSnapshot(input: unknown): MetricsSnapshot;
|
|
450
|
+
/**
|
|
451
|
+
* Human-readable descriptions of each scenario
|
|
452
|
+
*/
|
|
453
|
+
export declare const SCENARIO_DESCRIPTIONS: Record<ScenarioType, string>;
|
|
454
|
+
/**
|
|
455
|
+
* Get description for a scenario
|
|
456
|
+
*/
|
|
457
|
+
export declare function getScenarioDescription(scenario: ScenarioType): string;
|
|
458
|
+
//# sourceMappingURL=types.d.ts.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"types.d.ts","sourceRoot":"","sources":["../../src/benchmark/types.ts"],"names":[],"mappings":"AAAA;;;;;GAKG;AAEH,OAAO,EAAE,CAAC,EAAE,MAAM,KAAK,CAAC;AAMxB;;GAEG;AACH,eAAO,MAAM,gBAAgB;;;;;;EAM3B,CAAC;AAEH,MAAM,MAAM,YAAY,GAAG,CAAC,CAAC,KAAK,CAAC,OAAO,gBAAgB,CAAC,CAAC;AAE5D;;GAEG;AACH,eAAO,MAAM,kBAAkB;;;EAAgD,CAAC;AAEhF,MAAM,MAAM,cAAc,GAAG,CAAC,CAAC,KAAK,CAAC,OAAO,kBAAkB,CAAC,CAAC;AAEhE;;GAEG;AACH,eAAO,MAAM,UAAU;;;;EAIrB,CAAC;AAEH,MAAM,MAAM,MAAM,GAAG,CAAC,CAAC,KAAK,CAAC,OAAO,UAAU,CAAC,CAAC;AAMhD;;GAEG;AACH,eAAO,MAAM,qBAAqB;;;;;;;;;;;;;iBAkBhC,CAAC;AAEH,MAAM,MAAM,eAAe,GAAG,CAAC,CAAC,KAAK,CAAC,OAAO,qBAAqB,CAAC,CAAC;AAMpE;;GAEG;AACH,eAAO,MAAM,qBAAqB;;;;;;;;iBAqBhC,CAAC;AAEH,MAAM,MAAM,eAAe,GAAG,CAAC,CAAC,KAAK,CAAC,OAAO,qBAAqB,CAAC,CAAC;AAMpE;;GAEG;AACH,eAAO,MAAM,sBAAsB;;;;;;;;;iBAwBjC,CAAC;AAEH,MAAM,MAAM,gBAAgB,GAAG,CAAC,CAAC,KAAK,CAAC,OAAO,sBAAsB,CAAC,CAAC;AAMtE;;GAEG;AACH,eAAO,MAAM,oBAAoB;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;iBAkB/B,CAAC;AAEH,MAAM,MAAM,cAAc,GAAG,CAAC,CAAC,KAAK,CAAC,OAAO,oBAAoB,CAAC,CAAC;AAMlE;;GAEG;AACH,eAAO,MAAM,iBAAiB;;;;;;;iBAO5B,CAAC;AAEH,MAAM,MAAM,WAAW,GAAG,CAAC,CAAC,KAAK,CAAC,OAAO,iBAAiB,CAAC,CAAC;AAE5D;;GAEG;AACH,eAAO,MAAM,gBAAgB;;;;;;;;iBAG3B,CAAC;AAEH,MAAM,MAAM,UAAU,GAAG,CAAC,CAAC,KAAK,CAAC,OAAO,gBAAgB,CAAC,CAAC;AAE1D;;GAEG;AACH,eAAO,MAAM,2BAA2B;;;;;;;;;;;;;;;;;;;;;iBAetC,CAAC;AAEH,MAAM,MAAM,qBAAqB,GAAG,CAAC,CAAC,KAAK,CAAC,OAAO,2BAA2B,CAAC,CAAC;AAMhF;;GAEG;AACH,eAAO,MAAM,sBAAsB;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;iBAejC,CAAC;AAEH,MAAM,MAAM,gBAAgB,GAAG,CAAC,CAAC,KAAK,CAAC,OAAO,sBAAsB,CAAC,CAAC;AAMtE;;GAEG;AACH,eAAO,MAAM,sBAAsB;;;;;;;iBAsB/B,CAAC;AAEL,MAAM,MAAM,gBAAgB,GAAG,CAAC,CAAC,KAAK,CAAC,OAAO,sBAAsB,CAAC,CAAC;AAMtE;;GAEG;AACH,eAAO,MAAM,qBAAqB;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;iBA8B9B,CAAC;AAEL,MAAM,MAAM,eAAe,GAAG,CAAC,CAAC,KAAK,CAAC,OAAO,qBAAqB,CAAC,CAAC;AAMpE;;GAEG;AACH,wBAAgB,mBAAmB,CACjC,SAAS,CAAC,EAAE,OAAO,CAAC,eAAe,CAAC,GACnC,eAAe,CAwBjB;AAED;;;GAGG;AACH,wBAAgB,uBAAuB,CAAC,KAAK,EAAE,OAAO,GAAG,eAAe,CAEvE;AAED;;;GAGG;AACH,wBAAgB,uBAAuB,CAAC,KAAK,EAAE,OAAO,GAAG,eAAe,CAEvE;AAED;;;GAGG;AACH,wBAAgB,uBAAuB,CAAC,KAAK,EAAE,OAAO,GAAG,eAAe,CAEvE;AAMD;;GAEG;AACH,eAAO,MAAM,qBAAqB,EAAE,MAAM,CAAC,YAAY,EAAE,MAAM,CAW9D,CAAC;AAEF;;GAEG;AACH,wBAAgB,sBAAsB,CAAC,QAAQ,EAAE,YAAY,GAAG,MAAM,CAErE"}
|