2026.10.6: load-balance weights, upstream concurrency, model specs, key usage limits (core v0.63.0)

package.json

@@ -2,7 +2,7 @@
   "name": "thinkwatch-lite",
   "private": true,
   "description": "Desktop app for a local AI API gateway on macOS, Windows and Linux",
-  "version": "2026.10.5",
+  "version": "2026.10.6",
   "type": "module",
   "packageManager": "pnpm@11.13.0",
   "scripts": {

release-notes/2026.10.6.md

@@ -0,0 +1,30 @@
+**Upgrade notes:**
+- The bundled core is now 0.63.0, which speaks control-plane protocol 39. A remote server has to run core 0.63.0 too: this version does not connect to 0.62.0, and 2026.10.5 and earlier do not connect to 0.63.0.
+- Request history is kept.
+- A configuration edited by hand that lists an upstream twice in a strategy group, or a name that is not an upstream, no longer loads; safe mode names the group and the member.
+- A key's concurrency limit now counts streamed answers until they end, so a key at its limit waits more often than before.
+
+**Strategy groups:**
+- Load-balance groups take a weight per member, from 1 to 100, and a choice of how to distribute: by ratio, by speed, by reliability, or both. Speed is the time to first token, reliability the recent success rate; both adjust the weights, and an upstream without enough measurements counts as average.
+- Ongoing conversations stay on their upstream and count toward its share. The route test shows each member's weight, speed, success rate and share, and predicts the next upstream.
+
+**Upstreams:**
+- An upstream can have a concurrency limit. A conversation in progress waits for a free slot; other requests go to the next upstream, and when every upstream is full the request waits, then fails with a clear message.
+- A model's context window and output limit can be set by hand under "Specs…" in the upstream's model list, for relays the price table does not know or gets wrong. A manual value is marked and replaces the price table's.
+- Check-up findings (a different model name in answers, input reported high or low, few cache reads) now appear on the upstream's row; the separate Check-up tab is gone.
+
+**Failover settings:**
+- "Move to the next upstream when the start times out" (off by default): a streamed answer that has sent nothing within the wait moves to the next upstream, if one can take it. The last upstream keeps waiting.
+- "Wait for a free slot at most" (30 seconds by default) is the total time a request waits for a full upstream or a key's per-minute or per-hour limit.
+
+**Keys:**
+- A key can have usage limits: requests, tokens or cost (USD) per minute, hour, day, week or month, all of which must hold. A token limit can count cache reads. Day, week and month reset at local midnight, on Monday and on the 1st.
+- The key dialog shows each limit's usage and reset time, and lists the models the key can use that have no price; they count as $0. Keys at a limit are marked in the list, and a notification comes at 80% and when a limit is reached.
+
+**Traffic:**
+- The attempt chain shows an upstream given up for a slow start, with the input tokens it may have billed, upstreams skipped at their concurrency limit, and time spent waiting for a slot.
+- Requests refused by a key's usage limit or because every upstream was full are labelled as such.
+- On a Responses WebSocket connection, each turn is now a request of its own, with its usage and cost.
+
+**Also:**
+- The menu bar's menu follows the appearance chosen in the app instead of the menu bar's tint, so it no longer opens light in Dark Mode over a bright wallpaper.

scripts/shots/core/en/keys.json

@@ -3,6 +3,7 @@
     "allow": null,
     "default": true,
     "key": "tw-m9EXAMPLEj7ak",
+    "limits": [],
     "max_concurrent": null,
     "name": "default",
     "route": null
@@ -11,6 +12,7 @@
     "allow": null,
     "client": "claude-code",
     "key": "tw-5aEXAMPLEqmg8",
+    "limits": [],
     "max_concurrent": null,
     "name": "claude-code",
     "route": null
@@ -19,6 +21,7 @@
     "allow": null,
     "client": "codex",
     "key": "tw-h5EXAMPLEyrqn",
+    "limits": [],
     "max_concurrent": 4,
     "name": "codex",
     "route": "codex"
@@ -30,6 +33,7 @@
     ],
     "client": "cursor",
     "key": "tw-e6EXAMPLE6td6",
+    "limits": [],
     "max_concurrent": null,
     "name": "cursor",
     "route": "cursor"

scripts/shots/core/en/overview.json

@@ -26,6 +26,7 @@
       "allow": null,
       "default": true,
       "key": "tw-m9…j7ak",
+      "limits": [],
       "max_concurrent": null,
       "name": "default",
       "route": null
@@ -34,6 +35,7 @@
       "allow": null,
       "client": "claude-code",
       "key": "tw-5a…qmg8",
+      "limits": [],
       "max_concurrent": null,
       "name": "claude-code",
       "route": null
@@ -42,6 +44,7 @@
       "allow": null,
       "client": "codex",
       "key": "tw-h5…yrqn",
+      "limits": [],
       "max_concurrent": 4,
       "name": "codex",
       "route": "codex"
@@ -53,6 +56,7 @@
       ],
       "client": "cursor",
       "key": "tw-e6…6td6",
+      "limits": [],
       "max_concurrent": null,
       "name": "cursor",
       "route": "cursor"
@@ -63,32 +67,39 @@
   "failover": {
     "failures_to_pause": 3,
     "max_pause_secs": 600,
+    "next_on_slow_start": false,
     "no_balance_pause_secs": 1800,
     "pause_secs": 60,
     "quota_pause_secs": 3600,
     "rate_limit_max_pause_secs": 3600,
+    "slot_wait_secs": 30,
     "stream_start_wait_secs": 15
   },
   "groups": [
     {
+      "balance_by": "weights",
       "builtin": false,
       "kind": "fallback",
       "name": "main",
       "providers": [
         "anthropic",
         "relay"
-      ]
+      ],
+      "weights": {}
     },
     {
+      "balance_by": "weights",
       "builtin": false,
       "kind": "cheapest",
       "name": "budget",
       "providers": [
         "anthropic",
         "relay"
-      ]
+      ],
+      "weights": {}
     },
     {
+      "balance_by": "weights",
       "builtin": true,
       "kind": "fallback",
       "name": "__all__",
@@ -100,7 +111,8 @@
         "deepseek",
         "gemini",
         "ollama"
-      ]
+      ],
+      "weights": {}
     }
   ],
   "listen": {

scripts/shots/core/en/status.json

@@ -14,5 +14,5 @@
     "reachable": []
   },
   "uptime_secs": 0,
-  "version": "0.61.0"
+  "version": "0.62.0"
 }

scripts/shots/core/oracle.sh

@@ -8,15 +8,16 @@
 # 钉住的 core 之后。截图页的配置类数据（上游、密钥、路由、安全规则、试算）
 # 直接用这几份文件，所以它们必须是 core 真的答出来的，不是照着样子手写的。
 #
-# 用的是 src-tauri/Cargo.toml 钉住的那个 tag：从检出里 `git archive` 一份到临时
-# 目录，放进 oracle.rs 跑一次。**不改那个检出。**要先在那边 `git fetch --tags`。
+# 用的是 src-tauri/Cargo.toml 钉住的那个 tag（core 还没发版时临时钉的 rev 也认）：从检出里
+# `git archive` 一份到临时目录，放进 oracle.rs 跑一次。**不改那个检出。**要先在那边
+# `git fetch --tags`。
 set -euo pipefail
 
 core=${1:?用法：oracle.sh <thinkwatch-core 的本地检出>}
 here=$(cd "$(dirname "$0")" && pwd)
 root=$(cd "$here/../../.." && pwd)
-tag=$(sed -n 's/^tw-api = .*tag = "\([^"]*\)".*/\1/p' "$root/src-tauri/Cargo.toml")
-[ -n "$tag" ] || { echo "src-tauri/Cargo.toml 里找不到 tw-api 的 tag" >&2; exit 1; }
+tag=$(sed -nE 's/^tw-api = .*(tag|rev) = "([^"]*)".*/\2/p' "$root/src-tauri/Cargo.toml")
+[ -n "$tag" ] || { echo "src-tauri/Cargo.toml 里找不到 tw-api 的 tag 或 rev" >&2; exit 1; }
 
 src=$(mktemp -d)
 trap 'rm -rf "$src"' EXIT

scripts/shots/core/zh/keys.json

@@ -3,6 +3,7 @@
     "allow": null,
     "default": true,
     "key": "tw-m9EXAMPLEj7ak",
+    "limits": [],
     "max_concurrent": null,
     "name": "default",
     "route": null
@@ -11,6 +12,7 @@
     "allow": null,
     "client": "claude-code",
     "key": "tw-5aEXAMPLEqmg8",
+    "limits": [],
     "max_concurrent": null,
     "name": "claude-code",
     "route": null
@@ -19,6 +21,7 @@
     "allow": null,
     "client": "codex",
     "key": "tw-h5EXAMPLEyrqn",
+    "limits": [],
     "max_concurrent": 4,
     "name": "codex",
     "route": "codex"
@@ -30,6 +33,7 @@
     ],
     "client": "cursor",
     "key": "tw-e6EXAMPLE6td6",
+    "limits": [],
     "max_concurrent": null,
     "name": "cursor",
     "route": "cursor"

scripts/shots/core/zh/overview.json

@@ -26,6 +26,7 @@
       "allow": null,
       "default": true,
       "key": "tw-m9…j7ak",
+      "limits": [],
       "max_concurrent": null,
       "name": "default",
       "route": null
@@ -34,6 +35,7 @@
       "allow": null,
       "client": "claude-code",
       "key": "tw-5a…qmg8",
+      "limits": [],
       "max_concurrent": null,
       "name": "claude-code",
       "route": null
@@ -42,6 +44,7 @@
       "allow": null,
       "client": "codex",
       "key": "tw-h5…yrqn",
+      "limits": [],
       "max_concurrent": 4,
       "name": "codex",
       "route": "codex"
@@ -53,6 +56,7 @@
       ],
       "client": "cursor",
       "key": "tw-e6…6td6",
+      "limits": [],
       "max_concurrent": null,
       "name": "cursor",
       "route": "cursor"
@@ -63,32 +67,39 @@
   "failover": {
     "failures_to_pause": 3,
     "max_pause_secs": 600,
+    "next_on_slow_start": false,
     "no_balance_pause_secs": 1800,
     "pause_secs": 60,
     "quota_pause_secs": 3600,
     "rate_limit_max_pause_secs": 3600,
+    "slot_wait_secs": 30,
     "stream_start_wait_secs": 15
   },
   "groups": [
     {
+      "balance_by": "weights",
       "builtin": false,
       "kind": "fallback",
       "name": "主力",
       "providers": [
         "anthropic",
         "relay"
-      ]
+      ],
+      "weights": {}
     },
     {
+      "balance_by": "weights",
       "builtin": false,
       "kind": "cheapest",
       "name": "低价",
       "providers": [
         "anthropic",
         "relay"
-      ]
+      ],
+      "weights": {}
     },
     {
+      "balance_by": "weights",
       "builtin": true,
       "kind": "fallback",
       "name": "__all__",
@@ -100,7 +111,8 @@
         "deepseek",
         "gemini",
         "ollama"
-      ]
+      ],
+      "weights": {}
     }
   ],
   "listen": {

scripts/shots/core/zh/status.json

@@ -14,5 +14,5 @@
     "reachable": []
   },
   "uptime_secs": 0,
-  "version": "0.61.0"
+  "version": "0.62.0"
 }

scripts/shots/mock/core.ts

@@ -121,6 +121,7 @@ export const CORE: { [N in WebviewEndpoint]: Handler<N> } = {
   UpdateProvider: refuse,
   DeleteProvider: refuse,
   ProviderModels: (_req, [name]) => providerModels(name!) ?? notFound(`Upstream ${name}`),
+  SetModelSpec: refuse,
   RefreshProviderModels: (_req, [name]) => providerModels(name!) ?? notFound(`Upstream ${name}`),
   RefreshStaleModels: () => ({ providers: [] }),