MarkBase Admin
ac75faa0cc
CI / build-and-test (push) Has been cancelled
- E4B-MarkBase model (42 layers, 4.4GB) loaded successfully - All Phase 1-6 tests passed (model loading, forward pass, vision/audio towers, token generation, performance) - All stress tests passed (5/5 in 127.6s) - Concurrent inference - Memory stress (67.5 tok/s, 0 NaN) - Continuous generation - Batch processing - Long-running stability - Swift Metal inference engine with multimodal support
89 lines
4.1 KiB
Swift
89 lines
4.1 KiB
Swift
import XCTest
|
|
@testable import MarkBase
|
|
|
|
final class TextModelLoadTest: XCTestCase {
|
|
|
|
func testE4BTextLoad() throws {
|
|
print("\n═══════════════════════════════════════")
|
|
print(" E4B TEXT Model Load Test")
|
|
print("═══════════════════════════════════════\n")
|
|
|
|
let modelDir = "/Users/accusys/MarkBaseEngine/models/E4B-MarkBase"
|
|
|
|
print("Step 1: Load TEXT model only...")
|
|
let engine = try MarkBaseEngine(autoCompile: true)
|
|
let loadStart = Date()
|
|
let textModel = try E4BModel(modelDir: modelDir, engine: engine, maxContextLength: 256)
|
|
let loadTime = Date().timeIntervalSince(loadStart) * 1000
|
|
|
|
print(" ✓ TEXT model loaded in \(loadTime) ms")
|
|
print(" Hidden: \(textModel.hiddenSize)")
|
|
print(" Layers: \(textModel.numHiddenLayers)")
|
|
|
|
print("\nStep 2: Generate 10 tokens...")
|
|
var tokens: [Int] = [2] // BOS
|
|
let genStart = Date()
|
|
for _ in 0..<10 {
|
|
let logits = try textModel.forward(tokenId: tokens.last!, position: tokens.count - 1)
|
|
var maxIdx = 0
|
|
var maxLogit = logits[0]
|
|
for i in 1..<logits.count {
|
|
if logits[i] > maxLogit {
|
|
maxLogit = logits[i]
|
|
maxIdx = i
|
|
}
|
|
}
|
|
tokens.append(maxIdx)
|
|
}
|
|
let genTime = Date().timeIntervalSince(genStart) * 1000
|
|
print(" ✓ Generated 10 tokens in \(genTime) ms")
|
|
print(" Tokens: \(tokens.suffix(10))")
|
|
|
|
print("\n═══════════════════════════════════════")
|
|
print("✓ E4B TEXT test passed")
|
|
print(" Load: \(loadTime) ms, Gen: \(genTime) ms")
|
|
print("═══════════════════════════════════════\n")
|
|
}
|
|
|
|
func test12BTextLoad() throws {
|
|
print("\n═══════════════════════════════════════")
|
|
print(" 12B TEXT Model Load Test")
|
|
print("═══════════════════════════════════════\n")
|
|
|
|
let modelDir = "/Users/accusys/.cache/huggingface/hub/models--mlx-community--gemma-4-12B-it-4bit/snapshots/73bcf09092aa277861d5a191b989b666f7f32e8f"
|
|
|
|
print("Step 1: Load TEXT model only...")
|
|
let engine = try MarkBaseEngine(autoCompile: true)
|
|
let loadStart = Date()
|
|
let textModel = try E4BModel(modelDir: modelDir, engine: engine, maxContextLength: 256)
|
|
let loadTime = Date().timeIntervalSince(loadStart) * 1000
|
|
|
|
print(" ✓ TEXT model loaded in \(loadTime) ms")
|
|
print(" Hidden: \(textModel.hiddenSize)")
|
|
print(" Layers: \(textModel.numHiddenLayers)")
|
|
|
|
print("\nStep 2: Generate 10 tokens...")
|
|
var tokens: [Int] = [2] // BOS
|
|
let genStart = Date()
|
|
for _ in 0..<10 {
|
|
let logits = try textModel.forward(tokenId: tokens.last!, position: tokens.count - 1)
|
|
var maxIdx = 0
|
|
var maxLogit = logits[0]
|
|
for i in 1..<logits.count {
|
|
if logits[i] > maxLogit {
|
|
maxLogit = logits[i]
|
|
maxIdx = i
|
|
}
|
|
}
|
|
tokens.append(maxIdx)
|
|
}
|
|
let genTime = Date().timeIntervalSince(genStart) * 1000
|
|
print(" ✓ Generated 10 tokens in \(genTime) ms")
|
|
print(" Tokens: \(tokens.suffix(10))")
|
|
|
|
print("\n═══════════════════════════════════════")
|
|
print("✓ 12B TEXT test passed")
|
|
print(" Load: \(loadTime) ms, Gen: \(genTime) ms")
|
|
print("═══════════════════════════════════════\n")
|
|
}
|
|
} |