copilot-custom-endpoint 1.5.2 → 1.5.4
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 +40 -27
- package/docs/models/kimi.md +5 -5
- package/docs/models/mimo.md +4 -2
- package/docs/models/qwen.md +1 -1
- package/docs/pricing.md +25 -24
- package/package.json +2 -2
package/README.md
CHANGED
|
@@ -92,39 +92,52 @@ npx copilot-custom-endpoint clean # Remove debug_log/
|
|
|
92
92
|
|
|
93
93
|
All prices are **USD per 1M tokens** (non-cached). 1 AI credit = $0.01. To convert to AI credits, multiply by 100 (e.g., $5.00/1M = 500 credits/1M). Session cost assumes ~10K input + ~2K output tokens per turn, 50 turns.
|
|
94
94
|
|
|
95
|
-
| Model
|
|
96
|
-
|
|
|
97
|
-
| **DeepSeek V4 Flash**
|
|
98
|
-
|
|
|
99
|
-
| **
|
|
100
|
-
| **
|
|
101
|
-
| **
|
|
102
|
-
| **
|
|
103
|
-
| **
|
|
104
|
-
| **
|
|
105
|
-
| **
|
|
106
|
-
| **
|
|
107
|
-
| **
|
|
108
|
-
| **
|
|
109
|
-
| **
|
|
110
|
-
| **
|
|
111
|
-
| **
|
|
112
|
-
| **GPT-5.
|
|
113
|
-
| **
|
|
114
|
-
| **
|
|
115
|
-
| **
|
|
116
|
-
| **
|
|
117
|
-
| **
|
|
118
|
-
|
|
|
119
|
-
| **
|
|
120
|
-
|
|
|
121
|
-
|
|
|
95
|
+
| Model | Provider | Cost per intelligence | Intelligence Score | Input (per 1M) | Cached input | Output (per 1M) | Est. session | Context window |
|
|
96
|
+
| --------------------- | --------- | --------------------- | ------------------ | ----------------------------- | ----------------------------- | ----------------------------- | ------------ | -------------- |
|
|
97
|
+
| **DeepSeek V4 Flash** | DeepSeek | **~$0.0025** | **40.3** | $0.14 | $0.0028 | $0.28 | ~$0.10 | 1M |
|
|
98
|
+
| MiMo V2 Flash | Xiaomi | **~$0.0035** | **23.0** ³ | $0.10 | $0.01 | $0.30 | ~$0.08 | 256K |
|
|
99
|
+
| **MiniMax M3** | MiniMax | **~$0.0061** | **44.4** | $0.60 (≤512K) / $1.20 (>512K) | $0.12 (≤512K) / $0.24 (>512K) | $2.40 (≤512K) / $4.80 (>512K) | ~$0.27 | 1M |
|
|
100
|
+
| **DeepSeek V4 Pro** | DeepSeek | **~$0.0068** | **44.3** | $0.435 | $0.003625 | $0.87 | ~$0.30 | 1M |
|
|
101
|
+
| **Qwen 3.7 Plus** | DashScope | **~$0.0092** | **39.0** | $0.40 (≤256K) / $1.20 (>256K) | — | $1.60 (≤256K) / $4.80 (>256K) | ~$0.36 | 1M |
|
|
102
|
+
| **MiMo V2.5 Pro** | Xiaomi | **~$0.019** | **42.2** | $1.00 | $0.20 | $3.00 | ~$0.80 | 1M |
|
|
103
|
+
| **Gemini 3 Flash** | Google | **~$0.020** | **27.0** ³ | $0.50 | $0.05 | $3.00 | ~$0.55 | 173K |
|
|
104
|
+
| **Kimi K2.6** | Moonshot | **~$0.021** | **42.8** | $0.95 | $0.16 | $4.00 | ~$0.88 | 256K |
|
|
105
|
+
| **Kimi K2.7 Code** | Moonshot | **~$0.021** | **42.0** | $0.95 | $0.19 | $4.00 | ~$0.88 | 262K |
|
|
106
|
+
| **GPT-5.4 mini** | OpenAI | **~$0.021** | **40.0** | $0.75 | $0.075 | $4.50 | ~$0.83 | 400K |
|
|
107
|
+
| **GLM 5.1** | Z.ai | **~$0.028** | **40.2** | $1.40 | $0.26 | $4.40 | ~$1.14 | 200K |
|
|
108
|
+
| **GLM 5V Turbo** | Z.ai | **~$0.029** | **34.0** ³ | $1.20 | $0.24 | $4.00 | ~$1.00 | 200K |
|
|
109
|
+
| **Gemini 3.5 Flash** | Google | **~$0.033** | **50.2** | $1.50 | $0.15 | $9.00 | ~$1.65 | 1M |
|
|
110
|
+
| **Qwen 3.7 Max** | DashScope | **~$0.043** | **46.0** | $2.50 (≤1M) | — | $7.50 (≤1M) | ~$2.00 | 1M |
|
|
111
|
+
| **Gemini 3.1 Pro** | Google | **~$0.047** | **46.5** | $2.00¹ | $0.20 | $12.00¹ | ~$2.20 | 1M |
|
|
112
|
+
| **GPT-5.3-Codex** | OpenAI | **~$0.052** | **44.0** ³ | $1.75 | $0.175 | $14.00 | ~$2.28 | 400K |
|
|
113
|
+
| **GPT-5.4** | OpenAI | **~$0.054** | **51.4** | $2.50 | $0.25 | $15.00 | ~$2.75 | 1M |
|
|
114
|
+
| **Claude Sonnet 4.6** | Anthropic | **~$0.064** | **47.2** | $3.00 | $0.30 | $15.00 | ~$3.00 | 1M |
|
|
115
|
+
| **Claude Opus 4.8** | Anthropic | **~$0.090** | **55.7** | $5.00 | $0.50 | $25.00 | ~$5.00 | 1M |
|
|
116
|
+
| **Claude Opus 4.7** | Anthropic | **~$0.093** | **53.5** | $5.00 | $0.50 | $25.00 | ~$5.00 | 1M |
|
|
117
|
+
| **GPT-5.5** | OpenAI | **~$0.10** | **54.8** | $5.00 | $0.50 | $30.00 | ~$5.50 | 1M |
|
|
118
|
+
| Raptor mini | GitHub | — | — | $0.25 | $0.025 | $2.00 | ~$0.33 | 264K |
|
|
119
|
+
| **MiMo V2.5** | Xiaomi | — | — | $0.40 | $0.08 | $2.00 | ~$0.40 | 1M |
|
|
120
|
+
| MAI-Code-1-Flash ² | Microsoft | — | — | $0.75 | $0.075 | $4.50 | ~$0.83 | — |
|
|
121
|
+
| Claude Haiku 4.5 | Anthropic | — | — | $1.00 | $0.10 | $5.00 | ~$1.00 | 160K |
|
|
122
122
|
|
|
123
123
|
¹ Gemini 3.1 Pro pricing applies to prompts ≤200K tokens.
|
|
124
|
+
|
|
124
125
|
² MAI-Code-1-Flash is a continuously improving model — performance and behavior may evolve over time as new checkpoints are released.
|
|
125
126
|
|
|
127
|
+
³ Score is an **estimate** from Artificial Analysis (labelled "independent evaluation forthcoming"). Not a confirmed run of the full evaluation suite.
|
|
128
|
+
|
|
126
129
|
For footnotes, sources, and detailed notes (cache behavior, tiered pricing, free quotas) see [docs/pricing.md](docs/pricing.md). For a copy-paste config containing **all providers at once**, see [docs/example-config.md](docs/example-config.md).
|
|
127
130
|
|
|
131
|
+
> **👤 Personal picks** —
|
|
132
|
+
>
|
|
133
|
+
> For serious coding work, **MiniMax M3** is my top recommendation:
|
|
134
|
+
>
|
|
135
|
+
> - Highest intelligence score (44.4) among affordable models
|
|
136
|
+
> - Only ~$0.27/session with vision + 1M context
|
|
137
|
+
> - No local proxy required — works direct
|
|
138
|
+
>
|
|
139
|
+
> If you need something even cheaper for simpler tasks, **DeepSeek V4 Flash** (~$0.10/session, 40.3 score) is a solid fallback. MiMo V2 Flash is the absolute cheapest (~$0.08/session), but its low score (23.0) makes it unreliable for multi-step agentic changes.
|
|
140
|
+
|
|
128
141
|
## Companion tools
|
|
129
142
|
|
|
130
143
|
> **ℹ️ These are third-party tools — not built into the custom endpoints or proxies in this repo.** Each one must be installed, configured, and (where applicable) billed for **separately**, directly with its own provider. Nothing here is bundled, proxied, or auto-configured by `copilot-custom-endpoint` or the per-model setups above. The entries below are just pointers to tools the author has found useful alongside the model configs.
|
package/docs/models/kimi.md
CHANGED
|
@@ -37,7 +37,7 @@
|
|
|
37
37
|
1. **Start the proxy** — choose one:
|
|
38
38
|
- `npm run proxy:kimi` (from the repo root)
|
|
39
39
|
- `npx copilot-custom-endpoint kimi` (standalone, no clone needed)
|
|
40
|
-
- `npx copilot-custom-endpoint` (also starts the Qwen
|
|
40
|
+
- `npx copilot-custom-endpoint` (also starts the Qwen and MiMo proxies concurrently)
|
|
41
41
|
2. **Edit `chatLanguageModels.json`** — add the Kimi block from [Setup](#setup) below.
|
|
42
42
|
3. **Set your Moonshot API key** via the Command Palette → **Chat: Manage Language Models**.
|
|
43
43
|
4. **Restart VS Code** and pick "Kimi K2.6" or "Kimi K2.7 Code" in the chat picker.
|
|
@@ -188,10 +188,10 @@ All can be set in a `.env` file at the repo root (both proxies `import 'dotenv/c
|
|
|
188
188
|
|
|
189
189
|
For the cross-provider comparison, see [docs/pricing.md](../pricing.md). Kimi models on the **Moonshot direct platform**:
|
|
190
190
|
|
|
191
|
-
| Model | Input
|
|
192
|
-
| ---------------- |
|
|
193
|
-
| `kimi-k2.6` | $0.
|
|
194
|
-
| `kimi-k2.7-code` | $0.
|
|
191
|
+
| Model | Input (cache miss) | Cached input (cache hit) | Output |
|
|
192
|
+
| ---------------- | ------------------ | ------------------------ | ---------- |
|
|
193
|
+
| `kimi-k2.6` | $0.95 / 1M | $0.16 / 1M | $4.00 / 1M |
|
|
194
|
+
| `kimi-k2.7-code` | $0.95 / 1M | $0.19 / 1M | $4.00 / 1M |
|
|
195
195
|
|
|
196
196
|
> **K2.7:** No non-thinking mode — always-thinking. Cached input pricing applies.
|
|
197
197
|
> Via DashScope, K2.6 is also available at $0.89 / 1M input and $3.71 / 1M output (same model, regional pricing).
|
package/docs/models/mimo.md
CHANGED
|
@@ -36,8 +36,10 @@
|
|
|
36
36
|
|
|
37
37
|
The `proxy/mimo-proxy.mjs` provides dynamic thinking suppression: reasoning stays ON in plain chat but turns OFF automatically when tools are invoked.
|
|
38
38
|
|
|
39
|
-
- `npm run proxy:mimo` (
|
|
40
|
-
- `
|
|
39
|
+
- `npm run proxy:mimo` (from the repo root)
|
|
40
|
+
- `npx copilot-custom-endpoint mimo` (standalone, no clone needed)
|
|
41
|
+
- `npx copilot-custom-endpoint` (also starts the Kimi and Qwen proxies concurrently)
|
|
42
|
+
- `node proxy/mimo-proxy.mjs --help` — prints all supported environment variables and defaults
|
|
41
43
|
|
|
42
44
|
When using the proxy, update your model URLs to `http://127.0.0.1:3459/v1/chat/completions` and **remove** `thinking` from `requestBody`. The proxy handles it dynamically.
|
|
43
45
|
|
package/docs/models/qwen.md
CHANGED
|
@@ -29,7 +29,7 @@
|
|
|
29
29
|
1. **Start the proxy** — choose one:
|
|
30
30
|
- `npm run proxy:qwen` (from the repo root)
|
|
31
31
|
- `npx copilot-custom-endpoint qwen` (standalone, no clone needed)
|
|
32
|
-
- `npx copilot-custom-endpoint` (also starts the Kimi
|
|
32
|
+
- `npx copilot-custom-endpoint` (also starts the Kimi and MiMo proxies concurrently)
|
|
33
33
|
2. **Edit `chatLanguageModels.json`** — use the proxy-path block from [Setup § Proxy](#proxy-path) below.
|
|
34
34
|
3. **Set your DashScope API key** via the Language Models UI.
|
|
35
35
|
4. **Restart VS Code.** Reasoning will be visible in plain chat and suppressed on tool turns.
|
package/docs/pricing.md
CHANGED
|
@@ -18,40 +18,41 @@
|
|
|
18
18
|
|
|
19
19
|
All prices below are in **USD per 1M tokens** (non-cached). To convert to AI credits, multiply by 100 (e.g., $5.00/1M = 500 credits/1M).
|
|
20
20
|
|
|
21
|
-
All models are listed together below, sorted by
|
|
21
|
+
All models are listed together below, sorted by Cost per intelligence ascending (lower is better). Models without an Intelligence score are ordered by estimated session cost ascending. Session cost assumes ~10K input + ~2K output tokens per turn, 50 turns.
|
|
22
22
|
|
|
23
|
-
| Model | Provider |
|
|
23
|
+
| Model | Provider | Cost per intelligence | Intelligence Score | Input (per 1M) | Cached input | Output (per 1M) | Est. session | Context window |
|
|
24
24
|
| ---------------------- | --------- | --------------------- | ------------------ | ----------------------------- | ----------------------------- | ----------------------------- | ------------ | -------------- |
|
|
25
|
-
| **DeepSeek V4 Flash** | DeepSeek | **
|
|
26
|
-
| **
|
|
27
|
-
| **
|
|
28
|
-
| **
|
|
29
|
-
| **
|
|
30
|
-
| **
|
|
31
|
-
| **
|
|
32
|
-
| **
|
|
33
|
-
| **
|
|
34
|
-
| **
|
|
35
|
-
| **
|
|
36
|
-
| **
|
|
37
|
-
| **
|
|
38
|
-
| **
|
|
39
|
-
| **
|
|
40
|
-
| **GPT-5.
|
|
41
|
-
| **
|
|
25
|
+
| **DeepSeek V4 Flash** | DeepSeek | **~$0.0025** | **40.3** | $0.14 | $0.0028 | $0.28 | ~$0.10 | 1M |
|
|
26
|
+
| **MiMo V2 Flash** | Xiaomi | **~$0.0035** | **23.0** ³ | $0.10 | $0.01 | $0.30 | ~$0.08 | 256K |
|
|
27
|
+
| **MiniMax M3** | MiniMax | **~$0.0061** | **44.4** | $0.60 (≤512K) / $1.20 (>512K) | $0.12 (≤512K) / $0.24 (>512K) | $2.40 (≤512K) / $4.80 (>512K) | ~$0.27 | 1M |
|
|
28
|
+
| **DeepSeek V4 Pro** | DeepSeek | **~$0.0068** | **44.3** | $0.435 | $0.003625 | $0.87 | ~$0.30 | 1M |
|
|
29
|
+
| **Qwen 3.7 Plus** | DashScope | **~$0.0092** | **39.0** | $0.40 (≤256K) / $1.20 (>256K) | — | $1.60 (≤256K) / $4.80 (>256K) | ~$0.36 | 1M |
|
|
30
|
+
| **MiMo V2.5 Pro** | Xiaomi | **~$0.019** | **42.2** | $1.00 | $0.20 | $3.00 | ~$0.80 | 1M |
|
|
31
|
+
| **Gemini 3 Flash** | Google | **~$0.020** | **27.0** ³ | $0.50 | $0.05 | $3.00 | ~$0.55 | 173K |
|
|
32
|
+
| **Kimi K2.6** | Moonshot | **~$0.021** | **42.8** | $0.95 | $0.16 | $4.00 | ~$0.88 | 256K |
|
|
33
|
+
| **Kimi K2.7 Code** | Moonshot | **~$0.021** | **42.0** | $0.95 | $0.19 | $4.00 | ~$0.88 | 262K |
|
|
34
|
+
| **GPT-5.4 mini** | OpenAI | **~$0.021** | **40.0** | $0.75 | $0.075 | $4.50 | ~$0.83 | 400K |
|
|
35
|
+
| **GLM 5.1** | Z.ai | **~$0.028** | **40.2** | $1.40 | $0.26 | $4.40 | ~$1.14 | 200K |
|
|
36
|
+
| **GLM 5V Turbo** | Z.ai | **~$0.029** | **34.0** ³ | $1.20 | $0.24 | $4.00 | ~$1.00 | 200K |
|
|
37
|
+
| **Gemini 3.5 Flash** | Google | **~$0.033** | **50.2** | $1.50 | $0.15 | $9.00 | ~$1.65 | 1M |
|
|
38
|
+
| **Qwen 3.7 Max** | DashScope | **~$0.043** | **46.0** | $2.50 (≤1M) | — | $7.50 (≤1M) | ~$2.00 | 1M |
|
|
39
|
+
| **Gemini 3.1 Pro** | Google | **~$0.047** | **46.5** | $2.00¹ | $0.20 | $12.00¹ | ~$2.20 | 1M |
|
|
40
|
+
| **GPT-5.3-Codex** | OpenAI | **~$0.052** | **44.0** ³ | $1.75 | $0.175 | $14.00 | ~$2.28 | 400K |
|
|
41
|
+
| **GPT-5.4** | OpenAI | **~$0.054** | **51.4** | $2.50 | $0.25 | $15.00 | ~$2.75 | 1M |
|
|
42
|
+
| **Claude Sonnet 4.6** | Anthropic | **~$0.064** | **47.2** | $3.00 | $0.30 | $15.00 | ~$3.00 | 1M |
|
|
43
|
+
| **Claude Opus 4.8** | Anthropic | **~$0.090** | **55.7** | $5.00 | $0.50 | $25.00 | ~$5.00 | 1M |
|
|
44
|
+
| **Claude Opus 4.7** | Anthropic | **~$0.093** | **53.5** | $5.00 | $0.50 | $25.00 | ~$5.00 | 1M |
|
|
45
|
+
| **GPT-5.5** | OpenAI | **~$0.10** | **54.8** | $5.00 | $0.50 | $30.00 | ~$5.50 | 1M |
|
|
42
46
|
| **Raptor mini** | GitHub | — | — | $0.25 | $0.025 | $2.00 | ~$0.33 | 264K |
|
|
43
47
|
| **MiMo V2.5** | Xiaomi | — | — | $0.40 | $0.08 | $2.00 | ~$0.40 | 1M |
|
|
44
|
-
| **Gemini 3 Flash** | Google | — | — | $0.50 | $0.05 | $3.00 | ~$0.55 | 173K |
|
|
45
48
|
| **MAI-Code-1-Flash** ² | Microsoft | — | — | $0.75 | $0.075 | $4.50 | ~$0.83 | — |
|
|
46
|
-
| **Kimi K2.7 Code** | Moonshot | — | — | $0.95 | $0.19 | $4.00 | ~$0.88 | 262K |
|
|
47
49
|
| **Claude Haiku 4.5** | Anthropic | — | — | $1.00 | $0.10 | $5.00 | ~$1.00 | 160K |
|
|
48
|
-
| **GLM 5V Turbo** | Z.ai | — | — | $1.20 | $0.24 | $4.00 | ~$1.00 | 200K |
|
|
49
|
-
| **GPT-5.3-Codex** | OpenAI | — | — | $1.75 | $0.175 | $14.00 | ~$2.28 | 400K |
|
|
50
50
|
|
|
51
51
|
¹ Gemini 3.1 Pro pricing applies to prompts ≤200K tokens.
|
|
52
52
|
² MAI-Code-1-Flash is a continuously improving model — performance and behavior may evolve over time as new checkpoints are released.
|
|
53
|
+
³ Score is an **estimate** from Artificial Analysis (labelled "independent evaluation forthcoming"). Not a confirmed run of the full evaluation suite.
|
|
53
54
|
|
|
54
|
-
|
|
55
|
+
Cost per intelligence = estimated session cost ÷ Intelligence Index score. Session cost assumes ~10K input + ~2K output tokens per turn, 50 turns.
|
|
55
56
|
|
|
56
57
|
> **Notes:**
|
|
57
58
|
>
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "copilot-custom-endpoint",
|
|
3
|
-
"version": "1.5.
|
|
3
|
+
"version": "1.5.4",
|
|
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
|
+
}
|