claude-multi 0.8.2 → 0.10.0

package/CHANGELOG.md

@@ -5,6 +5,26 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [0.10.0] - 2026-06-13
+
+### Changed
+- **GLM template moved to a three-tier model mapping**: Opus slot now uses `glm-5.2[1m]`, sonnet uses `glm-5.1`, haiku uses `glm-5-turbo`. The previous template put GLM-5.1 in opus and reused GLM-5-Turbo for both sonnet and haiku, so there was no real middle tier.
+- GLM-5.2 is the new opus-tier model with a 1,000,000-token context window, exposed through the `[1m]` model-name suffix. GLM-5.1 and GLM-5-Turbo are 200K models.
+- Removed the global `CLAUDE_CODE_AUTO_COMPACT_WINDOW` (131072) and `CLAUDE_AUTOCOMPACT_PCT_OVERRIDE` overrides from the GLM template. A single global window cannot fit a 1M model and two 200K models at once, so each tier now carries its own size: the `[1m]` suffix for GLM-5.2, and Claude Code's default 200K assumption for the other two, which matches their real limit.
+- Updated GLM model and context-window references across the provider docs, FAQ, environment-variables guide, and glossary to reflect the three models.
+
+### Blog
+- Blog post: [GLM-5.2 for Claude Code: 1M Context and Three-Tier Model Mapping](https://claude-multi.hmziq.xyz/blog/glm-5-2-three-tier-coding-plan/)
+
+## [0.9.0] - 2026-06-12
+
+### Changed
+- **Kimi template updated with tiered model mapping**: Opus slot now uses `kimi-k2.7-code`, sonnet uses `kimi-k2.6`, haiku uses `kimi-k2.5`. Previous template used `kimi-k2.5` for all roles.
+- Kimi template context window corrected from 128K to 256K (matching actual model spec).
+
+### Blog
+- Blog post: [Kimi K2.7 for Claude Code: Three-Tier Model Mapping, Agentic Benchmarks](https://claude-multi.hmziq.xyz/blog/kimi-k27-three-tier-agentic-coding/)
+
 ## [0.8.2] - 2026-06-03
 
 ### Added

package/README.md

@@ -218,7 +218,7 @@ If the menu shows a `!` health hint, press `!` to see what's wrong and fix it.
 | DeepSeek | `api.deepseek.com` | `deepseek` |
 | Xiaomi MiMo | `api.xiaomimimo.com` | `mimo` |
 | Xiaomi MiMo (Token Plan) | `token-plan-cn.xiaomimimo.com` | `mimo-token` |
-| Moonshot Kimi | `api.moonshot.ai` | `kimi` |
+| Moonshot Kimi | `api.moonshot.ai` | `kimi-k2.7-code, kimi-k2.6, kimi-k2.5` |
 | Alibaba Qwen | `dashscope-intl.aliyuncs.com` | `qwen` |
 | Alibaba Qwen Coding Plan | `coding-intl.dashscope.aliyuncs.com` | `qwen-coding` |
 

package/build/cli.js

@@ -7381,24 +7381,22 @@ var init_templates = __esm(() => {
     glm: {
       name: "glm",
       displayName: "GLM Coding Plan",
-      description: "GLM-5.1 and GLM-5-Turbo via z.ai Coding Plan subscription (Anthropic endpoint is coding-plan-only)",
+      description: "GLM-5.2, GLM-5.1 and GLM-5-Turbo via z.ai Coding Plan subscription (Anthropic endpoint is coding-plan-only)",
       settings: {
         env: {
           ANTHROPIC_AUTH_TOKEN: "",
           ANTHROPIC_BASE_URL: "https://api.z.ai/api/anthropic",
           API_TIMEOUT_MS: "3000000",
           ANTHROPIC_DEFAULT_HAIKU_MODEL: "glm-5-turbo",
-          ANTHROPIC_DEFAULT_SONNET_MODEL: "glm-5-turbo",
-          ANTHROPIC_DEFAULT_OPUS_MODEL: "glm-5.1",
-          ANTHROPIC_MODEL: "glm-5.1",
+          ANTHROPIC_DEFAULT_SONNET_MODEL: "glm-5.1",
+          ANTHROPIC_DEFAULT_OPUS_MODEL: "glm-5.2[1m]",
+          ANTHROPIC_MODEL: "glm-5.2[1m]",
           ANTHROPIC_SMALL_FAST_MODEL: "glm-5-turbo",
           ENABLE_THINKING: "true",
           REASONING_EFFORT: "high",
           MAX_THINKING_TOKENS: "8000",
           ENABLE_STREAMING: "true",
-          MAX_OUTPUT_TOKENS: "64000",
-          CLAUDE_CODE_AUTO_COMPACT_WINDOW: "131072",
-          CLAUDE_AUTOCOMPACT_PCT_OVERRIDE: "75"
+          MAX_OUTPUT_TOKENS: "64000"
         },
         includeCoAuthoredBy: false,
         alwaysThinkingEnabled: false
@@ -7499,22 +7497,22 @@ var init_templates = __esm(() => {
     kimi: {
       name: "kimi",
       displayName: "Moonshot Kimi",
-      description: "Kimi K2.5 models via moonshot.ai — pay-per-token only, no subscription plan",
+      description: "Kimi K2.7 Code, K2.6, K2.5 models via moonshot.ai — pay-per-token only, no subscription plan",
       settings: {
         env: {
           ANTHROPIC_AUTH_TOKEN: "",
           ANTHROPIC_BASE_URL: "https://api.moonshot.ai/anthropic",
           API_TIMEOUT_MS: "3000000",
-          ANTHROPIC_MODEL: "kimi-k2.5",
-          ANTHROPIC_DEFAULT_OPUS_MODEL: "kimi-k2.5",
-          ANTHROPIC_DEFAULT_SONNET_MODEL: "kimi-k2.5",
+          ANTHROPIC_MODEL: "kimi-k2.6",
+          ANTHROPIC_DEFAULT_OPUS_MODEL: "kimi-k2.7-code",
+          ANTHROPIC_DEFAULT_SONNET_MODEL: "kimi-k2.6",
           ANTHROPIC_DEFAULT_HAIKU_MODEL: "kimi-k2.5",
           ANTHROPIC_SMALL_FAST_MODEL: "kimi-k2.5",
           ENABLE_THINKING: "true",
           REASONING_EFFORT: "high",
           MAX_THINKING_TOKENS: "16000",
           MAX_OUTPUT_TOKENS: "65536",
-          CLAUDE_CODE_AUTO_COMPACT_WINDOW: "131072",
+          CLAUDE_CODE_AUTO_COMPACT_WINDOW: "262144",
           CLAUDE_AUTOCOMPACT_PCT_OVERRIDE: "75"
         },
         includeCoAuthoredBy: false,

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "claude-multi",
-  "version": "0.8.2",
+  "version": "0.10.0",
   "description": "Manage multiple Claude Code instances with different aliases and configurations",
   "type": "module",
   "bin": {