copilot-custom-endpoint 1.6.1 → 1.6.3

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/README.md CHANGED
@@ -21,7 +21,6 @@ That's it. No code, no servers to manage (unless the model specifically needs th
21
21
 
22
22
  | Model | Provider | Needs proxy? | Vision | Setup guide |
23
23
  | --------------------------- | --------- | ---------------------- | ------------ | -------------------------------------------------------------------------------------------------- |
24
- | **MiMo V2 Flash** | Xiaomi | Optional (recommended) | ❌ | [Setup](docs/models/mimo.md) |
25
24
  | **MiMo V2.5** | Xiaomi | Optional (recommended) | ✅ | [Setup](docs/models/mimo.md) |
26
25
  | **MiMo V2.5 Pro** | Xiaomi | Optional (recommended) | ❌ | [Setup](docs/models/mimo.md) |
27
26
  | **Kimi K2.7 Code / K2.6** | Moonshot | **Yes** | ✅ | [Setup](docs/models/kimi.md) |
@@ -115,7 +114,6 @@ All prices are **USD per 1M tokens** (non-cached). 1 AI credit = $0.01. To conve
115
114
  | **Claude Opus 4.8** | Anthropic | **~$0.090** | **55.7** | ~$5.00 | ✅ | 1M |
116
115
  | **Claude Opus 4.7** | Anthropic | **~$0.093** | **53.5** | ~$5.00 | ✅ | 1M |
117
116
  | **GPT-5.5** | OpenAI | **~$0.10** | **54.8** | ~$5.50 | ✅ | 1M |
118
- | **MiMo V2 Flash** | Xiaomi | — | — | ~$0.08 | ❌ | 256K |
119
117
  | **Raptor mini** | GitHub | — | — | ~$0.33 | ❌ | 264K |
120
118
  | **MiMo V2.5** | Xiaomi | — | — | ~$0.40 | ✅ | 1M |
121
119
  | **MAI-Code-1-Flash** ² | Microsoft | — | — | ~$0.83 | ❌ | — |
@@ -138,10 +136,15 @@ For footnotes, sources, and detailed notes (cache behavior, tiered pricing, free
138
136
  >
139
137
  > **DeepSeek V4 Pro** is a strong backup — nearly identical intelligence (44.3 vs 44.4) and cost (~$0.0068 vs ~$0.0061 per intelligence point), but noticeably faster in practice.
140
138
  >
139
+ > For **plan mode / architecture & design thinking**, **GLM 5.2** is worth the premium:
140
+ >
141
+ > - Highest intelligence score in the table (**51.0**) — best for reasoning-heavy planning
142
+ > - 1M lossless context, ideal for whole-codebase planning and large spec review
143
+ > - ~$1.14/session is ~4× MiniMax M3, so **not a daily driver for agent mode** — reserve it for planning, then hand implementation off to a cheaper model.
144
+ >
141
145
  > Cheaper alternatives for simpler tasks:
142
146
  >
143
147
  > - **DeepSeek V4 Flash** — ~$0.10/session, 40.3 score (solid fallback)
144
- > - **MiMo V2 Flash** — ~$0.08/session (cheapest, but unreliable for multi-step agentic work)
145
148
 
146
149
  ## Companion tools
147
150
 
@@ -100,20 +100,6 @@ Here's a complete, real-world `chatLanguageModels.json` that combines **the `cus
100
100
  "temperature": 1,
101
101
  "top_p": 0.95
102
102
  }
103
- },
104
- {
105
- "id": "mimo-v2-flash",
106
- "name": "MiMo V2 Flash (text)",
107
- "url": "http://127.0.0.1:3459/v1/chat/completions",
108
- "toolCalling": true,
109
- "vision": false,
110
- "streaming": true,
111
- "maxInputTokens": 262144,
112
- "maxOutputTokens": 65536,
113
- "requestBody": {
114
- "temperature": 0.3,
115
- "top_p": 0.95
116
- }
117
103
  }
118
104
  ]
119
105
  },
@@ -203,7 +189,7 @@ If you only need one provider, jump straight to its setup guide:
203
189
 
204
190
  - [Kimi K2.6 / K2.7 Code](kimi.md)
205
191
  - [Qwen 3.7 Plus / 3.7 Max](qwen.md)
206
- - [Xiaomi MiMo (V2.5 / V2.5 Pro / V2 Flash)](mimo.md)
192
+ - [Xiaomi MiMo (V2.5 / V2.5 Pro)](mimo.md)
207
193
  - [MiniMax M3](minimax.md)
208
194
  - [GLM (5.2 / 5.1 / 5V Turbo)](glm.md)
209
195
 
@@ -12,6 +12,18 @@
12
12
  | Upstream endpoint | `https://api.moonshot.ai/v1/chat/completions` |
13
13
  | Proxy endpoint | `http://127.0.0.1:3457/v1/chat/completions` |
14
14
 
15
+ ### Deprecated K2 model ids
16
+
17
+ Moonshot's official model catalog says the older `kimi-k2` preview and thinking variants were **officially discontinued on May 25, 2026** and are no longer maintained or supported. Deprecated ids:
18
+
19
+ - `kimi-k2-0905-preview`
20
+ - `kimi-k2-0711-preview`
21
+ - `kimi-k2-turbo-preview`
22
+ - `kimi-k2-thinking`
23
+ - `kimi-k2-thinking-turbo`
24
+
25
+ For current use, stick to `kimi-k2.6` or `kimi-k2.7-code`. As of June 24, 2026, Moonshot still lists `kimi-k2.5`, `kimi-k2.6`, `kimi-k2.7-code`, and `kimi-k2.7-code-highspeed` as active models.
26
+
15
27
  ### K2.6
16
28
 
17
29
  | Field | Value |
@@ -275,6 +287,7 @@ The model-level `requestBody.temperature = 1` override validated locally but was
275
287
 
276
288
  - VS Code custom endpoint docs: `https://code.visualstudio.com/docs/copilot/customization/language-models#_add-a-custom-endpoint-model`
277
289
  - Kimi docs index: `https://platform.kimi.ai/docs/llms.txt`
290
+ - Kimi model catalog (includes deprecated models): `https://platform.kimi.ai/docs/models.md`
278
291
  - Kimi chat completion docs: `https://platform.kimi.ai/docs/api/chat.md`
279
292
  - Kimi models list: `https://platform.kimi.ai/docs/api/list-models.md`
280
293
  - Kimi model parameter reference: `https://platform.kimi.ai/docs/api/models-overview.md`
@@ -1,6 +1,8 @@
1
1
  # Xiaomi MiMo — VS Code Custom Endpoint Setup Guide
2
2
 
3
3
  > **TL;DR:** MiMo works direct with static `thinking: { type: "disabled" }` in `requestBody`, or via `proxy/mimo-proxy.mjs` for dynamic thinking suppression (reasoning visible in plain chat, suppressed in tool loops). Static suppression is simpler; the proxy lets you see reasoning in non-agent chats.
4
+ >
5
+ > **June 2026 provider notice:** Xiaomi says the legacy ids `mimo-v2-pro` and `mimo-v2-omni` have already been auto-switched to V2.5 replacements and billed at V2.5-series pricing. Xiaomi's same notice also retires an older legacy chat alias and the `mimo-v2-tts` alias. Requests that still use retired legacy aliases are scheduled to start failing after **2026-06-30 00:00 Beijing time**. This guide now documents only the current V2.5 chat ids.
4
6
 
5
7
  ## At a Glance
6
8
 
@@ -9,8 +11,8 @@
9
11
  | Mode | **Direct** (proxy optional) |
10
12
  | Vision | ✅ Yes (`mimo-v2.5` only) |
11
13
  | Tool calling | ✅ Yes (with `thinking: disabled`) |
12
- | Context | 1M (V2.5 Pro / V2.5) / 256K (V2 Flash) |
13
- | Max output | 131072 (V2.5 Pro) / 32768 (V2.5) / 65536 (V2 Flash) |
14
+ | Context | 1M (V2.5 Pro / V2.5) |
15
+ | Max output | 131072 (V2.5 Pro) / 32768 (V2.5) |
14
16
  | Required `requestBody` | Direct: `thinking: { type: "disabled" }`<br>Proxy: none (proxy handles it) |
15
17
  | Endpoint | Direct: `https://api.xiaomimimo.com/v1/chat/completions`<br>Proxy: `http://127.0.0.1:3459/v1/chat/completions` |
16
18
 
@@ -20,9 +22,8 @@
20
22
  | --------------- | ------ | ------- | ------------------------------------------ |
21
23
  | `mimo-v2.5-pro` | ❌ | 1M | Flagship text-only — best for agentic work |
22
24
  | `mimo-v2.5` | ✅ | 1M | Omnimodal — text + image + video + audio |
23
- | `mimo-v2-flash` | ❌ | 256K | Fastest and cheapest — strong reasoning |
24
25
 
25
- > Legacy `mimo-v2-pro` and `mimo-v2-omni` auto-route to V2.5 (with V2.5 pricing) as of June 1, 2026, and will be fully deprecated by June 30, 2026. Use the V2.5 series.
26
+ > Official June 2026 notice: legacy pre-V2.5 chat aliases now auto-route to V2.5 replacements and become invalid after **2026-06-30 00:00 Beijing time**. This guide intentionally omits retired alias names so new configs only show supported ids.
26
27
 
27
28
  ## Quick Start
28
29
 
@@ -30,7 +31,7 @@
30
31
 
31
32
  1. **Edit `chatLanguageModels.json`** — add the MiMo block(s) from [Setup](#setup) below.
32
33
  2. **Set your `MIMO_API_KEY`** via Command Palette → **Chat: Manage Language Models**.
33
- 3. **Restart VS Code** and pick "MiMo V2.5 Pro", "MiMo V2.5", or "MiMo V2 Flash".
34
+ 3. **Restart VS Code** and pick "MiMo V2.5 Pro" or "MiMo V2.5".
34
35
 
35
36
  ### With optional proxy (dynamic thinking)
36
37
 
@@ -93,21 +94,6 @@ Config file location:
93
94
  "temperature": 1,
94
95
  "top_p": 0.95
95
96
  }
96
- },
97
- {
98
- "id": "mimo-v2-flash",
99
- "name": "MiMo V2 Flash (text)",
100
- "url": "https://api.xiaomimimo.com/v1/chat/completions",
101
- "toolCalling": true,
102
- "vision": false,
103
- "streaming": true,
104
- "maxInputTokens": 262144,
105
- "maxOutputTokens": 65536,
106
- "requestBody": {
107
- "thinking": { "type": "disabled" },
108
- "temperature": 0.3,
109
- "top_p": 0.95
110
- }
111
97
  }
112
98
  ]
113
99
  }
@@ -149,20 +135,6 @@ Config file location:
149
135
  "temperature": 1,
150
136
  "top_p": 0.95
151
137
  }
152
- },
153
- {
154
- "id": "mimo-v2-flash",
155
- "name": "MiMo V2 Flash (text)",
156
- "url": "http://127.0.0.1:3459/v1/chat/completions",
157
- "toolCalling": true,
158
- "vision": false,
159
- "streaming": true,
160
- "maxInputTokens": 262144,
161
- "maxOutputTokens": 65536,
162
- "requestBody": {
163
- "temperature": 0.3,
164
- "top_p": 0.95
165
- }
166
138
  }
167
139
  ]
168
140
  }
@@ -188,6 +160,20 @@ Token Plan subscribers use different base URLs and `tp-` prefixed keys:
188
160
 
189
161
  > Pay-as-you-go keys are `sk-…`; Token Plan keys are `tp-…`. The endpoint to use depends on which key you set.
190
162
 
163
+ ### 4. Legacy id retirement schedule
164
+
165
+ From Xiaomi's June 2026 deprecation notice:
166
+
167
+ | Legacy id | Replacement | Status on 2026-06-24 | Final cutoff |
168
+ | -------------- | --------------- | -------------------------------------------- | ----------------------------- |
169
+ | `mimo-v2-pro` | `mimo-v2.5-pro` | Auto-switched | 2026-06-30 00:00 Beijing time |
170
+ | `mimo-v2-omni` | `mimo-v2.5` | Auto-switched | 2026-06-30 00:00 Beijing time |
171
+ | `mimo-v2-tts` | `mimo-v2.5-tts` | Auto-switch on 2026-06-25 00:00 Beijing time | 2026-06-30 00:00 Beijing time |
172
+
173
+ Xiaomi's notice also includes an additional retired legacy chat alias that auto-switches to `mimo-v2.5`; it is intentionally omitted here so the examples only show the ids you should still configure.
174
+
175
+ For VS Code custom endpoint chat usage, configure the V2.5 ids directly rather than relying on retiring aliases.
176
+
191
177
  ## Configuration Reference
192
178
 
193
179
  ### Sampling parameters
@@ -208,7 +194,6 @@ MiMo accepts `temperature` in `[0, 1.5]` and `top_p` in `[0.01, 1.0]`.
208
194
  | Model | API default `thinking.type` | API default `temperature` |
209
195
  | ---------------------------- | --------------------------- | -------------------------- |
210
196
  | `mimo-v2.5-pro`, `mimo-v2.5` | `enabled` | `1.0` (locked in thinking) |
211
- | `mimo-v2-flash` | `disabled` | `0.3` (customizable) |
212
197
 
213
198
  When thinking is enabled, responses include a `reasoning_content` field alongside `content` and `tool_calls`.
214
199
 
@@ -229,7 +214,7 @@ When thinking is enabled, responses include a `reasoning_content` field alongsid
229
214
  | Symptom | Likely cause | Fix |
230
215
  | ------------------------------------------ | -------------------------------------------------------- | ---------------------------------------------------------------------------------------------- |
231
216
  | HTTP 400 on the second turn of a tool loop | `reasoning_content` missing in history (thinking on) | Add `thinking: { type: "disabled" }` to `requestBody`, or use the proxy (`npm run proxy:mimo`) |
232
- | Vision request returns an error | Used `mimo-v2.5-pro` or `mimo-v2-flash` (text-only) | Use `mimo-v2.5` for vision |
217
+ | Vision request returns an error | Used `mimo-v2.5-pro` (text-only) | Use `mimo-v2.5` for vision |
233
218
  | Custom `tool_choice` ignored | MiMo only honors `"auto"` | Stick to `auto` |
234
219
  | 401 Unauthorized | Wrong key, or Token Plan URL used with pay-as-you-go key | Match key prefix (`sk-` vs `tp-`) to the endpoint |
235
220
  | 429 rate-limited | Concurrent sessions exceeded 100 RPM / 10M TPM | Reduce concurrent agent sessions |
@@ -242,7 +227,6 @@ For the cross-provider comparison, see [docs/pricing.md](../pricing.md). Oversea
242
227
  | --------------- | ----------------- | ------------------ | ---------- |
243
228
  | `mimo-v2.5-pro` | $0.20 / 1M | $1.00 / 1M | $3.00 / 1M |
244
229
  | `mimo-v2.5` | $0.08 / 1M | $0.40 / 1M | $2.00 / 1M |
245
- | `mimo-v2-flash` | $0.01 / 1M | $0.10 / 1M | $0.30 / 1M |
246
230
 
247
231
  > Cache writing is currently free of charge (limited-time offer). MiMo also offers a Token Plan subscription with discounted rates and a free cache-writing promotion.
248
232
 
@@ -279,7 +263,6 @@ Static suppression (direct mode) remains a perfectly valid simpler alternative.
279
263
  | --------------- | ------------------ | ------------- | ------------------ | --------- |
280
264
  | `mimo-v2.5-pro` | — | 57.2% | 68.4% | — |
281
265
  | `mimo-v2.5` | — | 56.1% | — | — |
282
- | `mimo-v2-flash` | 73.4% | — | — | 94.1% |
283
266
 
284
267
  > `mimo-v2.5` additionally scores 87.7% on Video-MME and 62.3% on Claw-Eval Text.
285
268
 
@@ -294,12 +277,9 @@ Static suppression (direct mode) remains a perfectly valid simpler alternative.
294
277
 
295
278
  External API checks (curl):
296
279
 
297
- | Check | Model | Result |
298
- | ------------------ | --------------- | ------------------------------------------------------- |
299
- | Non-streaming chat | `mimo-v2-flash` | ✅ |
300
- | Streaming (SSE) | `mimo-v2-flash` | ✅ |
301
- | Non-streaming chat | `mimo-v2.5-pro` | ✅ |
302
- | Tool calling | `mimo-v2-flash` | ✅ — `finish_reason: "tool_calls"` with valid JSON args |
280
+ | Check | Model | Result |
281
+ | ------------------ | --------------- | ------ |
282
+ | Non-streaming chat | `mimo-v2.5-pro` | ✅ |
303
283
 
304
284
  ### Known risks
305
285
 
@@ -323,6 +303,5 @@ External API checks (curl):
323
303
  - AI Tools Integration: `https://platform.xiaomimimo.com/docs/en-US/integration/claude-code`
324
304
  - HuggingFace (MiMo-V2.5): `https://huggingface.co/XiaomiMiMo/MiMo-V2.5`
325
305
  - HuggingFace (MiMo-V2.5-Pro): `https://huggingface.co/XiaomiMiMo/MiMo-V2.5-Pro`
326
- - HuggingFace (MiMo-V2-Flash): `https://huggingface.co/XiaomiMiMo/MiMo-V2-Flash`
327
306
  - MiMo V2.5 Blog: `https://mimo.xiaomi.com/mimo-v2-5`
328
307
  - AI Studio (playground): `https://aistudio.xiaomimimo.com/`
package/docs/pricing.md CHANGED
@@ -43,7 +43,6 @@ All models are listed together below, sorted by Cost per intelligence ascending
43
43
  | **Claude Opus 4.8** | Anthropic | **~$0.090** | **55.7** | $5.00 | $0.50 | $25.00 | ~$5.00 | 1M |
44
44
  | **Claude Opus 4.7** | Anthropic | **~$0.093** | **53.5** | $5.00 | $0.50 | $25.00 | ~$5.00 | 1M |
45
45
  | **GPT-5.5** | OpenAI | **~$0.10** | **54.8** | $5.00 | $0.50 | $30.00 | ~$5.50 | 1M |
46
- | **MiMo V2 Flash** | Xiaomi | — | — | $0.10 | $0.01 | $0.30 | ~$0.08 | 256K |
47
46
  | **Raptor mini** | GitHub | — | — | $0.25 | $0.025 | $2.00 | ~$0.33 | 264K |
48
47
  | **MiMo V2.5** | Xiaomi | — | — | $0.40 | $0.08 | $2.00 | ~$0.40 | 1M |
49
48
  | **MAI-Code-1-Flash** ² | Microsoft | — | — | $0.75 | $0.075 | $4.50 | ~$0.83 | — |
@@ -62,7 +61,7 @@ Cost per intelligence = estimated session cost ÷ Intelligence Index score. Sess
62
61
  > **Notes:**
63
62
  >
64
63
  > - **DeepSeek V4** input pricing shown is the **cache miss** price. Cache hits are significantly cheaper ($0.0028/M for Flash, $0.003625/M for Pro).
65
- > - **MiMo** input pricing shown is the **cache miss** price. Cache hits are 5× cheaper for V2.5 Pro ($0.20/M) and V2.5 ($0.08/M), and 10× cheaper for V2 Flash ($0.01/M).
64
+ > - **MiMo** input pricing shown is the **cache miss** price. Cache hits are 5× cheaper for V2.5 Pro ($0.20/M) and V2.5 ($0.08/M).
66
65
  > - **Gemini 3 Flash** is priced at $0.50/MTok input (text/image/video) and $1.00/MTok input for audio.
67
66
  > - **Anthropic (Claude)** models also have a cache write cost ($6.25/MTok for Opus, $3.75/MTok for Sonnet, $1.25/MTok for Haiku). Opus 4.7+ use a new tokenizer that may use up to 35% more tokens for the same text.
68
67
  > - **OpenAI** models support cached input at 0.1× base input rate.
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "copilot-custom-endpoint",
3
- "version": "1.6.1",
3
+ "version": "1.6.3",
4
4
  "description": "Local proxies for VS Code Copilot custom endpoints — Kimi K2, Qwen 3.x & MiMo V2.5",
5
5
  "license": "MIT",
6
6
  "type": "module",
@@ -57,4 +57,4 @@
57
57
  "dependencies": {
58
58
  "dotenv": "^17.4.2"
59
59
  }
60
- }
60
+ }