mlboydaisuke commited on
Commit
51aac72
Β·
verified Β·
1 Parent(s): 75af886

gen-cards: regenerate Use-it block

Browse files
Files changed (1) hide show
  1. README.md +26 -10
README.md CHANGED
@@ -33,6 +33,8 @@ Part of the community Core AI model zoo: **https://github.com/john-rocky/coreai-
33
  <!-- gen-cards:use-it begin id=glm-4.7-flash (managed by scripts/gen-cards β€” edit cards.json / QuickStart.swift, not this block) -->
34
  ## Use it
35
 
 
 
36
  ⚑ **One line** β€” run the kit's task op on this model
37
  (`import CoreAIOps`; no session, no model plumbing, downloads on first use):
38
 
@@ -40,34 +42,48 @@ Part of the community Core AI model zoo: **https://github.com/john-rocky/coreai-
40
  let tldr = try await CoreAI.summarize(text, options: .model("glm-4.7-flash"))
41
  ```
42
 
43
- Every op, one shape β€” [Cookbook](https://github.com/john-rocky/coreai-kit/blob/main/docs/COOKBOOK.md).
44
 
45
- ▢️ **Run it (source)** β€” the [ChatDemo runner](https://github.com/john-rocky/coreai-kit/tree/main/Examples/ChatDemo)
46
  (GUI + CLI, one app for every chat model in the catalog):
47
 
48
  ```bash
49
- git clone https://github.com/john-rocky/coreai-kit
50
- open coreai-kit/Examples/ChatDemo/ChatDemo.xcodeproj
 
51
  # β†’ Run, then pick "GLM-4.7-Flash (MoE+MLA)" in the model picker
52
 
53
  # agents / headless (macOS):
54
  cd coreai-kit/Examples/ChatDemo
55
- swift run chat-cli --model glm-4.7-flash --prompt "What can you do, offline?"
56
  ```
57
 
 
 
58
  πŸ’» **Build with it** β€” complete; the glue is kit API, copy-paste runs:
59
 
60
  ```swift
61
  import CoreAIKit
62
 
63
- let chat = try await ChatSession(catalog: "glm-4.7-flash")
 
 
 
 
 
 
 
 
 
 
 
64
  let reply = try await chat.respond(to: prompt)
65
  // reply: the answer, generated fully on-device
66
  ```
67
 
68
- **When Apple's FoundationModels built-in model isn't enough, keep your session code and swap the model β€” one line.** CoreAIKit's [`KitLanguageModel`](https://github.com/john-rocky/coreai-kit#when-foundationmodels-isnt-enough) plugs this bundle into the same system `LanguageModelSession`; your `Tool`s, `@Generable` types and transcripts work unchanged, and capabilities (tool calling, guided generation) auto-detect per model.
69
 
70
- The take-home is [`Examples/ChatDemo/Sources/QuickStart.swift`](https://github.com/john-rocky/coreai-kit/blob/main/Examples/ChatDemo/Sources/QuickStart.swift)
71
  β€” this exact code as one typed function, no UI; the CLI is an argument shell over it, and
72
  the GUI drives the same `ChatSession` across turns for its transcript.
73
  Multi-turn? Hold the `ChatSession` and call `respond(to:)` per turn β€” it keeps the
@@ -75,10 +91,10 @@ conversation history; `streamResponse(to:)` yields tokens as they decode.
75
 
76
  **Integration checklist**
77
 
78
- - SPM: `https://github.com/john-rocky/coreai-kit` β†’ product **CoreAIKit**
79
  - Info.plist: none needed
80
  - Entitlements: none needed (macOS)
81
- - First run downloads the model β€” 30.0 GB (Mac) β€” then it loads from the
82
  local cache (Application Support; progress via the `downloadProgress` callback)
83
  - Measure in Release β€” Debug is ~3Γ— slower on per-token host work
84
  <!-- gen-cards:use-it end -->
 
33
  <!-- gen-cards:use-it begin id=glm-4.7-flash (managed by scripts/gen-cards β€” edit cards.json / QuickStart.swift, not this block) -->
34
  ## Use it
35
 
36
+ **New to Core AI? [Start with CoreAIKit 0.7.1](https://github.com/john-rocky/coreai-kit#readme).** Follow its requirements and first-run steps for `qwen3-0.6b`, then open the same release's [ChatDemo](https://github.com/john-rocky/coreai-kit/tree/0.7.1/Examples/ChatDemo). The README records the tested OS/SDK and download size; model and device coverage is stated per example.
37
+
38
  ⚑ **One line** β€” run the kit's task op on this model
39
  (`import CoreAIOps`; no session, no model plumbing, downloads on first use):
40
 
 
42
  let tldr = try await CoreAI.summarize(text, options: .model("glm-4.7-flash"))
43
  ```
44
 
45
+ Every op, one shape β€” [Cookbook](https://github.com/john-rocky/coreai-kit/blob/0.7.1/docs/COOKBOOK.md).
46
 
47
+ ▢️ **Run it (source)** β€” the [ChatDemo runner](https://github.com/john-rocky/coreai-kit/tree/0.7.1/Examples/ChatDemo)
48
  (GUI + CLI, one app for every chat model in the catalog):
49
 
50
  ```bash
51
+ git clone --branch 0.7.1 --depth 1 https://github.com/john-rocky/coreai-kit
52
+ export DEVELOPER_DIR=/Applications/Xcode-27.0.0-RC.app/Contents/Developer
53
+ open -a /Applications/Xcode-27.0.0-RC.app coreai-kit/Examples/ChatDemo/ChatDemo.xcodeproj
54
  # β†’ Run, then pick "GLM-4.7-Flash (MoE+MLA)" in the model picker
55
 
56
  # agents / headless (macOS):
57
  cd coreai-kit/Examples/ChatDemo
58
+ swift run -c release chat-cli --model glm-4.7-flash --prompt "What can you do, offline?"
59
  ```
60
 
61
+ Use Xcode build **27A266a** from the release's `.xcode-pin`; adjust the app path if your installation is named differently.
62
+
63
  πŸ’» **Build with it** β€” complete; the glue is kit API, copy-paste runs:
64
 
65
  ```swift
66
  import CoreAIKit
67
 
68
+ let id = "glm-4.7-flash"
69
+ let chat: ChatSession
70
+ if id == "qwen3-0.6b" {
71
+ // Freeze the release starter; other selections retain the live catalog's
72
+ // model-specific dispatch (including paired Gemma bundles).
73
+ guard let model = ModelCatalog.builtin.entry(id: id)?.modelID else {
74
+ throw CoreAIKitError.modelNotAvailableOnPlatform(id: id)
75
+ }
76
+ chat = try await ChatSession(model: model)
77
+ } else {
78
+ chat = try await ChatSession(catalog: "glm-4.7-flash")
79
+ }
80
  let reply = try await chat.respond(to: prompt)
81
  // reply: the answer, generated fully on-device
82
  ```
83
 
84
+ **When Apple's FoundationModels built-in model isn't enough, keep your session code and swap the model β€” one line.** CoreAIKit's [`KitLanguageModel`](https://github.com/john-rocky/coreai-kit#works-with-apples-foundationmodels-api) plugs supported chat bundles into the system `LanguageModelSession`. Tool calling depends on the model's dialect; guided generation also requires a compatible sequential engine. Check the linked support matrix before using either capability.
85
 
86
+ The take-home is [`Examples/ChatDemo/Sources/QuickStart.swift`](https://github.com/john-rocky/coreai-kit/blob/0.7.1/Examples/ChatDemo/Sources/QuickStart.swift)
87
  β€” this exact code as one typed function, no UI; the CLI is an argument shell over it, and
88
  the GUI drives the same `ChatSession` across turns for its transcript.
89
  Multi-turn? Hold the `ChatSession` and call `respond(to:)` per turn β€” it keeps the
 
91
 
92
  **Integration checklist**
93
 
94
+ - SPM: `https://github.com/john-rocky/coreai-kit` (exact **0.7.1**) β†’ product **CoreAIKit**
95
  - Info.plist: none needed
96
  - Entitlements: none needed (macOS)
97
+ - First run downloads the model β€” ~30,000 MB (Mac) β€” then it loads from the
98
  local cache (Application Support; progress via the `downloadProgress` callback)
99
  - Measure in Release β€” Debug is ~3Γ— slower on per-token host work
100
  <!-- gen-cards:use-it end -->