copilot-custom-endpoint 2.1.9 → 3.0.0
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 +74 -80
- package/docs/example-config.md +1 -1
- package/docs/models/qwen.md +64 -2
- package/docs/pricing.md +59 -46
- package/package.json +2 -2
- package/proxy/qwen-proxy.mjs +8 -1
package/README.md
CHANGED
|
@@ -20,20 +20,21 @@ That's it. No code, no servers to manage (unless the model specifically needs th
|
|
|
20
20
|
|
|
21
21
|
## Pick a model
|
|
22
22
|
|
|
23
|
-
| Model | Provider
|
|
24
|
-
| ------------------------------ |
|
|
25
|
-
| **Kimi K3 / K2.7 Code / K2.6** | Moonshot
|
|
26
|
-
| **MiMo V2.5** | Xiaomi
|
|
27
|
-
| **MiMo V2.5 Pro** | Xiaomi
|
|
28
|
-
| **Qwen 3.7 Plus** | DashScope
|
|
29
|
-
| **Qwen 3.7 Max** | DashScope
|
|
30
|
-
| **Qwen 3.8 Max**
|
|
31
|
-
| **
|
|
32
|
-
| **
|
|
33
|
-
| **GLM 5.3
|
|
34
|
-
| **GLM
|
|
35
|
-
| **
|
|
36
|
-
| **DeepSeek V4
|
|
23
|
+
| Model | Provider | Needs proxy? | Vision | Setup guide |
|
|
24
|
+
| ------------------------------ | ---------- | ----------------------- | ---------------------- | ------------------------------------------------------------------------------------------------------------------------------------------- |
|
|
25
|
+
| **Kimi K3 / K2.7 Code / K2.6** | Moonshot | **Yes** | ✅ | [Manual setup](docs/models/kimi.md) |
|
|
26
|
+
| **MiMo V2.5** | Xiaomi | Optional (recommended) | ✅ | [Extension](https://marketplace.visualstudio.com/items?itemName=sdmapvstool.xiaomimimo-for-copilot) ★ / [Manual setup](docs/models/mimo.md) |
|
|
27
|
+
| **MiMo V2.5 Pro** | Xiaomi | Optional (recommended) | ❌ | [Extension](https://marketplace.visualstudio.com/items?itemName=sdmapvstool.xiaomimimo-for-copilot) ★ / [Manual setup](docs/models/mimo.md) |
|
|
28
|
+
| **Qwen 3.7 Plus** | DashScope | Optional (recommended) | ✅ | [Manual setup](docs/models/qwen.md) |
|
|
29
|
+
| **Qwen 3.7 Max** | DashScope | Optional (recommended) | ❌ | [Manual setup](docs/models/qwen.md) |
|
|
30
|
+
| **Qwen 3.8 Max (0803)** | OpenRouter | Optional (experimental) | ✅ | [Snapshot/proxy setup](docs/models/qwen.md#optional-openrouter-snapshot-proxy) |
|
|
31
|
+
| **Qwen 3.8 Max (0902)** | DashScope | Optional (recommended) | ✅ | [Manual setup](docs/models/qwen.md) |
|
|
32
|
+
| **MiniMax M3** | MiniMax | No | ✅ | [Extension](https://github.com/tugudush/minimax-copilot) ★ / [Manual setup](docs/models/minimax.md) |
|
|
33
|
+
| **GLM 5.3 Flash** | Z.ai | No | ✅ | [Manual setup](docs/models/glm.md) |
|
|
34
|
+
| **GLM 5.3 / 5.2 / 5.1** | Z.ai | No | ✅ via extension proxy | [Manual setup](docs/models/glm.md) |
|
|
35
|
+
| **GLM 5V Turbo** | Z.ai | No | ✅ | [Manual setup](docs/models/glm.md) |
|
|
36
|
+
| **DeepSeek V4 Flash 0731** | DeepSeek | No (uses an extension) | ❌ | [Extension](https://marketplace.visualstudio.com/items?itemName=Vizards.deepseek-v4-for-copilot) / [setup notes](docs/models/deepseek.md) |
|
|
37
|
+
| **DeepSeek V4 Pro 0813** | DeepSeek | No (uses an extension) | ❌ | [Extension](https://marketplace.visualstudio.com/items?itemName=Vizards.deepseek-v4-for-copilot) / [setup notes](docs/models/deepseek.md) |
|
|
37
38
|
|
|
38
39
|
## Setup
|
|
39
40
|
|
|
@@ -107,50 +108,53 @@ npx copilot-custom-endpoint clean # Remove debug_log/
|
|
|
107
108
|
|
|
108
109
|
## Pricing snapshot
|
|
109
110
|
|
|
110
|
-
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. DeepSeek V4 estimates use peak rates; official off-peak rates are half.
|
|
111
|
-
|
|
112
|
-
| Model
|
|
113
|
-
|
|
|
114
|
-
| **GLM 5.3 Flash**
|
|
115
|
-
| **
|
|
116
|
-
| **
|
|
117
|
-
| **
|
|
118
|
-
| **
|
|
119
|
-
| **
|
|
120
|
-
| **
|
|
121
|
-
| **
|
|
122
|
-
| **Gemini 3.
|
|
123
|
-
| **
|
|
124
|
-
| **
|
|
125
|
-
| **
|
|
126
|
-
| **
|
|
127
|
-
| **
|
|
128
|
-
| **
|
|
129
|
-
| **
|
|
130
|
-
| **
|
|
131
|
-
| **
|
|
132
|
-
| **
|
|
133
|
-
| **
|
|
134
|
-
| **
|
|
135
|
-
| **
|
|
136
|
-
| **
|
|
137
|
-
| **
|
|
138
|
-
| **
|
|
139
|
-
| **
|
|
140
|
-
| **
|
|
141
|
-
| **
|
|
142
|
-
| **
|
|
143
|
-
| **
|
|
144
|
-
| **
|
|
145
|
-
| **
|
|
146
|
-
| **
|
|
147
|
-
| **
|
|
148
|
-
| **
|
|
149
|
-
| **GPT-5.
|
|
150
|
-
| **Claude
|
|
151
|
-
| **Claude
|
|
152
|
-
| **
|
|
153
|
-
| **
|
|
111
|
+
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. DeepSeek V4 estimates use peak rates; official off-peak rates are half. The current AA values come from OpenRouter's live model pages; **every pricing cell comes from first-party provider pricing or GitHub Copilot's official billing table, never OpenRouter.** OpenRouter lists Qwen 3.8 Max as two snapshots: unversioned **0803** at **53.4** and **0902** at **46.9**. Current corrections include OpenAI GPT-5.6 Sol at **$4 / $0.40 / $20**, Anthropic Sonnet 5 at **$2 / $0.20 / $10**, xAI Grok 4.5 cache reads at **$0.30**, and Alibaba's Qwen 3.7 list rates at **$2.50 / $0.25 / $7.50** (Max) and **$0.40 / $0.04 / $1.60** (Plus).
|
|
112
|
+
|
|
113
|
+
| Model | Provider | Cost per intelligence | Intelligence Score | Est. session | Vision | Context window |
|
|
114
|
+
| -------------------------- | --------- | --------------------- | ------------------ | ------------ | ------ | -------------- |
|
|
115
|
+
| **GLM 5.3 Flash** | Z.ai | **~$0.0028** | **46.2** | ~$0.13 | ✅ | 1M |
|
|
116
|
+
| **GPT-5.6 Luna** | OpenAI | **~$0.0051** | **43.4** | ~$0.22 | ✅ | 1M |
|
|
117
|
+
| **MiniMax M3** | MiniMax | **~$0.0076** | **35.7** | ~$0.27 | ✅ | 1M |
|
|
118
|
+
| **DeepSeek V4 Flash 0731** | DeepSeek | **~$0.0086** | **40.8** | ~$0.35 | ❌ | 1M |
|
|
119
|
+
| **MiMo V2.5 Pro** | Xiaomi | **~$0.0092** | **32.6** | ~$0.30 | ❌ | 1M |
|
|
120
|
+
| **MiniMax M3 Priority** | MiniMax | **~$0.0115** | **35.7** | ~$0.41 | ✅ | 1M |
|
|
121
|
+
| **Gemini 3.8 Flash** | Google | **~$0.0159** | **47.1** | ~$0.75 | ✅ | 1M |
|
|
122
|
+
| **Gemini 3.7 Flash** | Google | **~$0.0166** | **45.2** | ~$0.75 | ✅ | 1M |
|
|
123
|
+
| **Gemini 3.6 Flash** | Google | **~$0.0186** | **40.3** | ~$0.75 | ✅ | 1M |
|
|
124
|
+
| **GLM 5.3** | Z.ai | **~$0.0235** | **48.6** | ~$1.14 | ❌ | 1M |
|
|
125
|
+
| **DeepSeek V4 Pro 0813** | DeepSeek | **~$0.0252** | **42.1** | ~$1.06 | ❌ | 1M |
|
|
126
|
+
| **Qwen 3.8 Max (0803)** | DashScope | **~$0.0300** | **53.4** | ~$1.60 | ✅ | 1M |
|
|
127
|
+
| **Grok 4.6** | xAI | **~$0.0316** | **50.6** | ~$1.60 | ✅ | 500K |
|
|
128
|
+
| **Qwen 3.8 Max (0902)** | DashScope | **~$0.0341** | **46.9** | ~$1.60 | ✅ | 1M |
|
|
129
|
+
| **Grok 4.5** | xAI | **~$0.0352** | **45.5** | ~$1.60 | ✅ | 500K |
|
|
130
|
+
| **Claude Sonnet 5** | Anthropic | **~$0.0443** | **45.1** | ~$2.00 | ✅ | 1M |
|
|
131
|
+
| **GPT-5.6 Terra** | OpenAI | **~$0.0470** | **46.8** | ~$2.20 | ✅ | 1M |
|
|
132
|
+
| **Kimi K3** | Moonshot | **~$0.0598** | **50.2** | ~$3.00 | ✅ | 1M |
|
|
133
|
+
| **Gemini 3.1 Pro** | Google | **~$0.0599** | **36.7** | ~$2.20 | ✅ | 1M |
|
|
134
|
+
| **GPT-5.6 Sol** | OpenAI | **~$0.0780** | **51.3** | ~$4.00 | ✅ | 1M |
|
|
135
|
+
| **Claude Opus 5** | Anthropic | **~$0.0924** | **54.1** | ~$5.00 | ✅ | 1M |
|
|
136
|
+
| **Claude Fable 5.1** | Anthropic | **~$0.1761** | **56.8** | ~$10.00 | ✅ | 1M |
|
|
137
|
+
| **GPT-6 Astra** | OpenAI | **~$0.1828** | **54.7** | ~$10.00 | ✅ | 1M |
|
|
138
|
+
| **Claude Fable 5** | Anthropic | **~$0.1880** | **53.2** | ~$10.00 | ✅ | 1M |
|
|
139
|
+
| **MiMo V2.5** | Xiaomi | — | — | ~$0.10 | ✅ | 1M |
|
|
140
|
+
| **Qwen 3.7 Plus** | DashScope | — | — | ~$0.36 | ✅ | 1M |
|
|
141
|
+
| **Qwen 3.6 Plus** | DashScope | — | — | ~$0.55 | ✅ | 1M |
|
|
142
|
+
| **GPT-5.4 mini** | OpenAI | — | — | ~$0.83 | ❌ | 400K |
|
|
143
|
+
| **Kimi K2.6** | Moonshot | — | — | ~$0.88 | ✅ | 256K |
|
|
144
|
+
| **Kimi K2.7 Code** | Moonshot | — | — | ~$0.88 | ✅ | 262K |
|
|
145
|
+
| **GLM 5V Turbo** | Z.ai | — | — | ~$1.00 | ✅ | 200K |
|
|
146
|
+
| **GLM 5.1** | Z.ai | — | — | ~$1.14 | ❌ | 200K |
|
|
147
|
+
| **GLM 5.2** | Z.ai | — | — | ~$1.14 | ❌ | 1M |
|
|
148
|
+
| **Gemini 3.5 Flash** | Google | — | — | ~$1.65 | ✅ | 1M |
|
|
149
|
+
| **Qwen 3.7 Max** | DashScope | — | — | ~$2.00 | ❌ | 1M |
|
|
150
|
+
| **GPT-5.4** | OpenAI | — | — | ~$2.75 | ✅ | 1M |
|
|
151
|
+
| **Claude Sonnet 4.6** | Anthropic | — | — | ~$3.00 | ✅ | 1M |
|
|
152
|
+
| **Claude Opus 4.8** | Anthropic | — | — | ~$5.00 | ✅ | 1M |
|
|
153
|
+
| **Claude Opus 4.7** | Anthropic | — | — | ~$5.00 | ✅ | 1M |
|
|
154
|
+
| **GPT-5.5** | OpenAI | — | — | ~$5.50 | ✅ | 1M |
|
|
155
|
+
| **Gemini 3.8 Flash Cyber** | Google | — | — | — | — | — |
|
|
156
|
+
|
|
157
|
+
> **Historical footnotes:** The numbered notes below preserve release and pricing context from the prior September 5 snapshot. They are not the current AA ranking source; use the table above and its OpenRouter refresh note for current scores.
|
|
154
158
|
|
|
155
159
|
¹ Gemini 3.1 Pro pricing applies to prompts ≤200K tokens.
|
|
156
160
|
|
|
@@ -168,7 +172,7 @@ All prices are **USD per 1M tokens** (non-cached). 1 AI credit = $0.01. To conve
|
|
|
168
172
|
|
|
169
173
|
⁹ **Claude Opus 5** launched July 24, 2026. AA Intelligence Index score of **63.1** (#1 overall) confirmed by [Artificial Analysis](https://artificialanalysis.ai/models/claude-opus-5). Priced at $5.00 / $25.00 per MTok input/output (same as Opus 4.8). 1M context, text + image input, adaptive reasoning. Also available in Fast mode ($10/$50 per MTok input/output). Uses the newer Claude tokenizer (~30% more tokens than pre-4.7 models). See [Anthropic's Opus 5 announcement](https://www.anthropic.com/news/claude-opus-5).
|
|
170
174
|
|
|
171
|
-
¹⁰ **Qwen 3.8 Max
|
|
175
|
+
¹⁰ **Historical Qwen 3.8 Max AA record:** an earlier run reported **58.1** for the launch-era unversioned entry. Current OpenRouter rankings distinguish **Qwen 3.8 Max (0803)** at **53.4** from **Qwen 3.8 Max (0902)** at **46.9**; see the [benchmark table](docs/benchmarks.md) and [Qwen setup guide](docs/models/qwen.md#openrouter-snapshot-records).
|
|
172
176
|
|
|
173
177
|
¹¹ **Grok 4.5** (xAI, released July 8, 2026) is now a **GitHub Copilot native** model (GA, Versatile). AA Intelligence Index **55.8** (high), Coding **72.4**, Agentic **48.9**. 500K context, text + image input. Priced at $2.00 / $0.50 / $6.00 per 1M input/cached/output (75% cache discount; Copilot long-context >200K tier $4.00 / $1.00 / $12.00). See the [GitHub Copilot models & pricing page](https://docs.github.com/en/copilot/reference/copilot-billing/models-and-pricing).
|
|
174
178
|
|
|
@@ -194,33 +198,23 @@ All prices are **USD per 1M tokens** (non-cached). 1 AI credit = $0.01. To conve
|
|
|
194
198
|
|
|
195
199
|
²² **GPT-6 Astra** (released September 3, 2026) — OpenAI's frontier model, available as `gpt-6-astra` through the API and as a GitHub Copilot native model (GA, Powerful). AA Intelligence Index **61.2** (launch comparison, methodology v4.1.1); 1M context, text + image input, text output, and reasoning. Standard pricing is $10.00 / $1.00 / $50.00 per 1M input/cached/output tokens, for a modeled ~$10.00 session and ~$0.163 CPI. It is too new for Arena rankings. See the [GPT-6 Astra announcement](https://openai.com/index/gpt-6-astra/), [OpenAI API pricing](https://developers.openai.com/api/docs/pricing), and [OpenAI model documentation](https://developers.openai.com/api/docs/models/gpt-6-astra).
|
|
196
200
|
|
|
201
|
+
²³ **Qwen 3.8 Max (0803)** is the August 3, 2026 checkpoint behind OpenRouter's unversioned [`qwen/qwen3.8-max`](https://openrouter.ai/qwen/qwen3.8-max) row. It scores **53.4** in the September 7 snapshot and is retained as a comparison-only record because OpenRouter labels it superseded by 0902. The [Qwen setup guide](docs/models/qwen.md) explains why this frozen snapshot is not a second DashScope picker entry.
|
|
202
|
+
|
|
203
|
+
²⁴ **Qwen 3.8 Max (0902)** is the separate [`qwen/qwen3.8-max-0902`](https://openrouter.ai/qwen/qwen3.8-max-0902) snapshot, released September 4, 2026, with a current OpenRouter AA score of **46.9**. It shares Qwen Cloud's $2.00 / $0.25 / $6.00 rates, 1M context, and vision capability with 0803. The validated DashScope/proxy setup uses `qwen3.8-max`; the OpenRouter slug is not automatically interchangeable with that provider ID.
|
|
204
|
+
|
|
197
205
|
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).
|
|
198
206
|
|
|
199
207
|
> **👤 Personal picks** —
|
|
200
208
|
>
|
|
201
|
-
> For
|
|
202
|
-
>
|
|
203
|
-
> - **GPT-5.6 Luna** (Copilot native) — scores **52.3**, ~$0.22/session, vision-capable. It is cheaper than MiniMax M3 (45.4) and consumes fewer AI credits after OpenAI's July price cut.
|
|
204
|
-
> - **DeepSeek V4 Flash 0731** (extension) — scores **51.8** after the AA v4.1.1 benchmark refresh, ~$0.35/session at peak rates (~$0.18 off-peak). It is the lowest-cost text-only model above 50 in this table; pair it with a vision model when you need images.
|
|
205
|
-
> - **Gemini 3.7 Flash** (Copilot native, new August 13) — scores **56** at ~$0.75/session under promotional pricing. The best-value Google model in Copilot: it beats Gemini 3.5 Flash (52.0) and Grok 4.5 (55.8) on intelligence at roughly half the session cost, with 1M context, full multimodal input, and AA's fastest output speed. A strong upgrade pick if you use Google models.
|
|
206
|
-
> - **Gemini 3.8 Flash** (Gemini API, new September 2) — scores **58.7** at the same ~$0.75/session promotional price as 3.7 Flash, with +2.7 intelligence, 1M context, and full multimodal input. **Not yet Copilot native**, so it's a comparison row here rather than a config — and it runs very verbose (AA measured 120M output tokens on the Intelligence Index), so budget more output tokens than the score alone suggests.
|
|
207
|
-
>
|
|
208
|
-
> **MiniMax M3** (45.4, ~$0.27/session) is still the best **direct** custom-endpoint option — no proxy or extension, vision, 1M context, strong coding. Pick M3 when you want a plug-and-play custom endpoint with vision.
|
|
209
|
-
>
|
|
210
|
-
> For **plan mode / architecture & design thinking**, **GPT-6 Astra**, **Qwen 3.8 Max**, **Kimi K3**, **GLM 5.3**, **GLM 5.3 Flash**, **GPT-5.6 Luna**, and **DeepSeek V4 Flash 0731** are worth considering:
|
|
211
|
-
>
|
|
212
|
-
> - **GPT-6 Astra** — new OpenAI frontier model, scores **61.2** (#4 in this comparison), ~$10.00/session, with 1M context and vision. It is the strongest OpenAI option in the table, but its high token price makes it a reserve choice for the hardest planning and research tasks.
|
|
213
|
-
> - **Qwen 3.8 Max** — scores **58.1** (#5 Arena Text, the #2 custom-endpoint model) — the default **multimodal** planning pick. Near-flagship intelligence at ~$1.60/session, with 1M context and vision. Strong for whole-codebase planning, architecture, and large spec review.
|
|
214
|
-
> - **Kimi K3** — new flagship. Highest intelligence score among custom-endpoint models (**59.7**, #4 overall). Best for complex reasoning, long-horizon planning, and architecture decisions when you want the absolute best — but at ~$3.00/session it's ~2× Qwen 3.8 Max for +1.6 points, so **reserve it for hard problems**.
|
|
215
|
-
> - **GLM 5.3** — new flagship, scores **59.5** (now the #2 custom-endpoint model, just behind Kimi K3's 59.7) at ~$1.14/session. 1M context, text-only, always-thinking. It is the strongest text-only planning value above 59 and offers best-in-class agentic coding per Z.ai (50% gain over 5.2 on their Code Bench) — ideal for whole-codebase planning and large spec review. ~4× MiniMax M3, so **not a daily driver for agent mode** — reserve it for planning, then hand implementation off to a cheaper model.
|
|
216
|
-
> - **GLM 5.3 Flash** — new (Aug 26): scores **57.5** at just ~$0.13/session (list) — the **cheapest 57+ model** in the table and the first native multimodal GLM-5. Near-GLM-5.3 agentic coding (Z.ai: DeepSWE v1.1 63.4 vs 46.2) at ~9× less cost, with vision. A strong daily-driver planning + image-review pick until the 50% promo ends.
|
|
217
|
-
> - **GPT-5.6 Luna** (Copilot native) — scores **52.3** and is now the cheapest 50+ model. OpenAI cut its API price to $0.20/$1.20 per 1M input/output tokens; in Copilot, the equivalent benefit is lower AI-credit consumption, while your subscription price and monthly allowance stay unchanged. Ideal for high-quality planning on a budget.
|
|
218
|
-
> - **DeepSeek V4 Flash 0731** (extension) — scores **51.8** at ~$0.35/session at peak rates (~$0.18 off-peak) — the lower-cost planning pick. Text-only, so pair it with a vision model for design review, but it remains strong for reasoning-heavy planning.
|
|
209
|
+
> For current OpenRouter AA scores, **GPT-6 Astra** leads the OpenAI comparison benchmarks, while **Claude Fable 5.1** remains highest on the live AA composite. The two rankings measure different things.
|
|
219
210
|
>
|
|
220
|
-
>
|
|
211
|
+
> - **GLM 5.3 Flash** — current AA **46.2**, ~$0.13/session, vision-capable, and the best low-cost scored option in this table.
|
|
212
|
+
> - **GPT-5.6 Luna** — current AA **43.4**, ~$0.22/session, vision-capable, and the cheapest current OpenAI scored row.
|
|
213
|
+
> - **DeepSeek V4 Flash 0731** — current AA **40.8**, ~$0.35/session at peak direct rates, text-only and inexpensive for coding-heavy work.
|
|
214
|
+
> - **GPT-6 Astra** — current OpenRouter AA **54.7**, ~$10.00/session, while OpenAI reports leading results on Terminal-Bench 4.0, ARC-AGI-3, FrontierMath Tier 4, and AutomationBench. Use it for the hardest planning and research tasks.
|
|
215
|
+
> - **Qwen 3.8 Max (0902)** — current AA **46.9**, ~$1.60/session, 1M context, and vision-capable; a practical multimodal planning choice. The 0803 OpenRouter snapshot remains separately documented at **53.4**.
|
|
221
216
|
>
|
|
222
|
-
>
|
|
223
|
-
> - **MiMo V2.5** — ~$0.10/session, 38.0 score, vision-capable — the lowest modeled cost-per-intelligence option in the table
|
|
217
|
+
> **MiniMax M3** (current AA **35.7**, ~$0.27/session) remains the simplest direct custom-endpoint option: no proxy or extension, vision, 1M context, and tool calling.
|
|
224
218
|
|
|
225
219
|
## Companion tools
|
|
226
220
|
|
package/docs/example-config.md
CHANGED
|
@@ -6,7 +6,7 @@ Here's a complete, real-world `chatLanguageModels.json` that combines **the `cus
|
|
|
6
6
|
>
|
|
7
7
|
> **Note:** The `apiKey` fields are left as empty strings — set them via the **Chat: Manage Language Models** UI (Command Palette → right-click provider group → **Update API Key**). After you set a key via the UI, VS Code replaces the empty string with a `${input:chat.lm.secret.<id>}` secret reference.
|
|
8
8
|
>
|
|
9
|
-
> The live config points Qwen at the local proxy (`:3458`) and MiMo at the local proxy (`:3459`). When using a proxy, align the model `requestBody` overrides with the proxy's behavior: Qwen sends no `requestBody` (the proxy manages `enable_thinking` dynamically); MiMo sends only `temperature` and `top_p` (the proxy injects `thinking: {"type": "disabled"}` on tool turns and leaves it absent on plain chat).
|
|
9
|
+
> The live config points Qwen at the local proxy (`:3458`) and MiMo at the local proxy (`:3459`). When using a proxy, align the model `requestBody` overrides with the proxy's behavior: Qwen sends no `requestBody` (the proxy manages `enable_thinking` dynamically); MiMo sends only `temperature` and `top_p` (the proxy injects `thinking: {"type": "disabled"}` on tool turns and leaves it absent on plain chat). OpenRouter lists Qwen 3.8 Max as two frozen snapshots (`qwen/qwen3.8-max` = 0803 and `qwen/qwen3.8-max-0902` = 0902); this DashScope config intentionally uses the provider ID `qwen3.8-max` and does not pretend the OpenRouter slugs are interchangeable.
|
|
10
10
|
|
|
11
11
|
```json
|
|
12
12
|
[
|
package/docs/models/qwen.md
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
# Qwen (DashScope) — VS Code Custom Endpoint Setup Guide
|
|
2
2
|
|
|
3
|
-
> **TL;DR:** `qwen3.8-max` (vision), `qwen3.7-plus` (vision), and `qwen3.7-max` (text) work both direct and via the local proxy. The proxy gives you dynamic thinking suppression: reasoning stays ON in plain chat but turns OFF automatically when tools are invoked. The direct path is simpler if you don't need reasoning in chat.
|
|
3
|
+
> **TL;DR:** `qwen3.8-max` (vision), `qwen3.7-plus` (vision), and `qwen3.7-max` (text) work both direct and via the local proxy. OpenRouter currently lists two frozen Qwen 3.8 Max snapshots: the unversioned `qwen/qwen3.8-max` (0803) and `qwen/qwen3.8-max-0902`. The proxy forwards the upstream `model` value unchanged, so use the exact model ID supported by the endpoint you choose. The proxy gives you dynamic thinking suppression: reasoning stays ON in plain chat but turns OFF automatically when tools are invoked. The direct path is simpler if you don't need reasoning in chat.
|
|
4
4
|
|
|
5
5
|
## At a Glance
|
|
6
6
|
|
|
@@ -28,6 +28,68 @@
|
|
|
28
28
|
|
|
29
29
|
> The live `chatLanguageModels.json` points Qwen at the local proxy by default; the direct DashScope URL is shown below for users who prefer a static `enable_thinking: false` setup.
|
|
30
30
|
|
|
31
|
+
### OpenRouter snapshot records
|
|
32
|
+
|
|
33
|
+
These are separate OpenRouter catalog entries, not two picker labels for one model:
|
|
34
|
+
|
|
35
|
+
| OpenRouter slug | Snapshot | AA Intelligence Index | Notes |
|
|
36
|
+
| ----------------------- | -------- | --------------------- | --------------------------------------------------------------- |
|
|
37
|
+
| `qwen/qwen3.8-max` | 0803 | **53.4** | August 3 launch checkpoint; OpenRouter now labels it superseded |
|
|
38
|
+
| `qwen/qwen3.8-max-0902` | 0902 | **46.9** | September snapshot; released September 4 and listed separately |
|
|
39
|
+
|
|
40
|
+
The DashScope custom-endpoint snippets below use the provider model ID `qwen3.8-max`. Do not replace it with the OpenRouter `qwen/qwen3.8-max-0902` slug unless you also change the upstream URL to OpenRouter and have verified that path independently.
|
|
41
|
+
|
|
42
|
+
### Optional OpenRouter snapshot proxy
|
|
43
|
+
|
|
44
|
+
The proxy forwards the `model` field without mapping it, so it can also be pointed at OpenRouter for the two catalog snapshots. This is an **optional, unvalidated OpenRouter path**; it is separate from the validated DashScope setup above and requires an OpenRouter API key.
|
|
45
|
+
|
|
46
|
+
1. Add this to the repo-root `.env` file:
|
|
47
|
+
|
|
48
|
+
```dotenv
|
|
49
|
+
QWEN_UPSTREAM_URL=https://openrouter.ai/api/v1/chat/completions
|
|
50
|
+
```
|
|
51
|
+
|
|
52
|
+
2. Start the existing Qwen proxy:
|
|
53
|
+
|
|
54
|
+
```bash
|
|
55
|
+
npm run proxy:qwen
|
|
56
|
+
```
|
|
57
|
+
|
|
58
|
+
3. Add a separate provider group using the OpenRouter key and these model entries:
|
|
59
|
+
|
|
60
|
+
```json
|
|
61
|
+
{
|
|
62
|
+
"name": "Qwen (OpenRouter snapshots)",
|
|
63
|
+
"vendor": "customendpoint",
|
|
64
|
+
"apiKey": "",
|
|
65
|
+
"apiType": "chat-completions",
|
|
66
|
+
"models": [
|
|
67
|
+
{
|
|
68
|
+
"id": "qwen/qwen3.8-max",
|
|
69
|
+
"name": "Qwen 3.8 Max (0803)",
|
|
70
|
+
"url": "http://127.0.0.1:3458/v1/chat/completions",
|
|
71
|
+
"toolCalling": true,
|
|
72
|
+
"vision": true,
|
|
73
|
+
"streaming": true,
|
|
74
|
+
"maxInputTokens": 991000,
|
|
75
|
+
"maxOutputTokens": 131072
|
|
76
|
+
},
|
|
77
|
+
{
|
|
78
|
+
"id": "qwen/qwen3.8-max-0902",
|
|
79
|
+
"name": "Qwen 3.8 Max (0902)",
|
|
80
|
+
"url": "http://127.0.0.1:3458/v1/chat/completions",
|
|
81
|
+
"toolCalling": true,
|
|
82
|
+
"vision": true,
|
|
83
|
+
"streaming": true,
|
|
84
|
+
"maxInputTokens": 991000,
|
|
85
|
+
"maxOutputTokens": 131072
|
|
86
|
+
}
|
|
87
|
+
]
|
|
88
|
+
}
|
|
89
|
+
```
|
|
90
|
+
|
|
91
|
+
Set the API key through **Chat: Manage Language Models**, then verify the proxy with `curl http://127.0.0.1:3458/healthz`. Do not run this route at the same time as the default DashScope route unless each process uses a different `QWEN_PROXY_PORT` and log path.
|
|
92
|
+
|
|
31
93
|
## Quick Start
|
|
32
94
|
|
|
33
95
|
### With Proxy (recommended)
|
|
@@ -218,7 +280,7 @@ The proxy detects active tool use by examining the conversation state (a `"tool"
|
|
|
218
280
|
## Notes
|
|
219
281
|
|
|
220
282
|
- **Vision (`qwen3.8-max`, `qwen3.7-plus`)** uses OpenAI-compatible `content` array format. Base64 data URIs work reliably; external image URLs may fail if DashScope can't reach them. If a drag-and-drop image fails to load, providing the absolute file path (e.g. `c:\path\to\image.png`) in the prompt is a reliable workaround.
|
|
221
|
-
- **Qwen 3.8 reasoning:** `qwen3.8-max` enables reasoning by default and supports `reasoning_effort` values `low`, `medium`, and `xhigh` (default). The direct snippet disables reasoning for stable VS Code tool loops; the proxy leaves it on for plain chat and suppresses it when tool activity is detected.
|
|
283
|
+
- **Qwen 3.8 reasoning:** `qwen3.8-max` enables reasoning by default and supports `reasoning_effort` values `low`, `medium`, and `xhigh` (default). The direct snippet disables reasoning for stable VS Code tool loops; the proxy leaves it on for plain chat and suppresses it when tool activity is detected. The two OpenRouter snapshot slugs are catalog records; the benchmark difference does not mean the DashScope config should silently switch IDs.
|
|
222
284
|
- **Thinking trade-off:** Direct = thinking always off (loops stable, no reasoning visible). Proxy = thinking on in plain chat, off in tool turns.
|
|
223
285
|
- **`tool_choice` only supports `auto`** — don't override it (VS Code's default is `auto`).
|
|
224
286
|
|
package/docs/pricing.md
CHANGED
|
@@ -1,6 +1,10 @@
|
|
|
1
1
|
# Pricing
|
|
2
2
|
|
|
3
|
-
> **
|
|
3
|
+
> **Historical release timeline:** The long release log below is retained for provenance. The current table and the September 7 refresh note below are authoritative for present scores and estimates.
|
|
4
|
+
|
|
5
|
+
> **Updated:** September 7, 2026 — **Qwen 3.8 Max (0803)** and **Qwen 3.8 Max (0902)** are now separate rows (footnotes ²⁵–²⁶) because OpenRouter exposes them as distinct snapshots with AA scores **53.4** and **46.9**. Their Qwen Cloud rates and modeled session cost are unchanged. **GPT-6 Astra** added (footnote ²⁴): OpenAI's frontier model (released September 3, 2026), AA Intelligence Index **61.2**, standard pricing **$10 / $1 / $50** (input / cached / output), 1M context, and text + image input; it follows Claude Fable 5 in the CPI sort (~$0.163 vs ~$0.161). **Claude Fable 5.1** added (footnote ²³): Anthropic's new flagship (released September 1, 2026), AA Intelligence Index **65.7** — now the **top intelligence score** in the benchmarks table. Same $10 / $50 rates as Fable 5, but cached input cut to $0.25 (97.5% cache discount); it lands just above Fable 5 in the CPI sort (~$0.15 vs ~$0.161), still the second-most expensive scored row. **Gemini 3.8 Flash** added (footnote ²¹): Google's most intelligent Flash model (released September 2, 2026), AA Intelligence Index **58.7**, at the same promotional rates as 3.7 Flash ($0.75 / $0.075 / $3.75 through December 31) — it is **not yet GitHub Copilot native**, so it is priced from the Gemini API. **Gemini 3.8 Flash Cyber** added (footnote ²²): the cybersecurity-specialized sibling, restricted to Google's Fairwind Program — no public API pricing and no Artificial Analysis Intelligence Index run, so its row shows `—` at the bottom of the table. August 28: **GLM 5.3 Flash** added (footnote ²⁰): Z.ai's first native multimodal GLM-5 model (August 26, 2026), AA Intelligence Index **57.5**, launch pricing **$0.15 / $0.03 / $0.50** (50% off through September 9). It is now the **cheapest cost-per-intelligence row** in the table (~$0.0022). August 20: **DeepSeek V4 Flash 0731** and **DeepSeek V4 Pro 0813** refreshed (footnote ¹⁹): current builds, AA Intelligence Index **51.8** / **53.2**, and peak/off-peak API pricing effective August 16. Pro's official GA agent benchmark update is documented in [docs/models/deepseek.md](models/deepseek.md). August 19: **GLM 5.3** added (footnote ¹⁸): Z.ai's new flagship (August 18, 2026), AA Intelligence Index **59.5**, priced identically to GLM 5.2 ($1.40 / $0.26 / $4.40). August 17: **GLM 5.2** OpenRouter promo refreshed (footnote ⁴): NovitaAI cut to 48% off; Open Sail Research now cheapest at $0.50 / $3.15. August 15: **Claude Fable 5** and **Qwen 3.6 Plus** added (footnotes ¹⁶–¹⁷); **Grok 4.5 / 4.6** and **Claude Fable 5** are now **GitHub Copilot native** (footnotes ¹², ¹⁴, ¹⁶). **Grok 4.6** and **Gemini 3.7 Flash** added August 14 (footnotes ¹⁴–¹⁵). **Gemini 3.6 Flash promotional price cut** to $0.75 / $3.75 (footnote ⁹). **Qwen 3.7 Max / 3.7 Plus price cuts** (see footnote ¹³). OpenRouter is running limited-time promos on GPT-5.6 Terra/Luna (⁸), GLM 5.2 (⁴), GLM 5.1, MiMo V2.5/V2.5 Pro, and MiniMax M3.
|
|
6
|
+
|
|
7
|
+
> **Current refresh (September 7, 2026):** AA Intelligence Index values below use the current OpenRouter model pages. **Every pricing cell is checked against first-party provider pricing or GitHub Copilot's official billing table; OpenRouter is never used as a pricing source.** OpenRouter's unversioned `qwen/qwen3.8-max` is the **0803** checkpoint (**53.4**), while `qwen/qwen3.8-max-0902` is a separate **0902** snapshot (**46.9**); both use the same Qwen Cloud rates in this comparison. Several models no longer expose a current OpenRouter AA composite and are shown as `—` rather than retaining stale scores. Current corrections include OpenAI GPT-5.6 Sol at **$4 / $0.40 / $20**, Anthropic Sonnet 5 at **$2 / $0.20 / $10**, xAI Grok 4.5 cache reads at **$0.30**, and Alibaba's Qwen 3.7 list rates at **$2.50 / $0.25 / $7.50** (Max) and **$0.40 / $0.04 / $1.60** (Plus).
|
|
4
8
|
|
|
5
9
|
> **⏰ June 1, 2026 — GitHub Copilot switched to usage-based billing (AI Credits) today.**
|
|
6
10
|
>
|
|
@@ -18,52 +22,55 @@
|
|
|
18
22
|
>
|
|
19
23
|
> Code completions remain unlimited and **not** billed. Auto model selection gets a 10% discount.
|
|
20
24
|
|
|
21
|
-
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). DeepSeek V4 rows use peak rates for the comparison; its official off-peak rates are half.
|
|
25
|
+
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). DeepSeek V4 rows use peak rates for the comparison; its official off-peak rates are half. Provider promotions are shown separately in the notes; OpenRouter's effective, batch, or free-provider prices are excluded.
|
|
22
26
|
|
|
23
27
|
All models are listed together below, sorted by Cost per intelligence ascending (lower is better). Models without a Cost per intelligence are ordered by estimated session cost ascending. Session cost assumes ~10K input + ~2K output tokens per turn, 50 turns.
|
|
24
28
|
|
|
25
|
-
| Model
|
|
26
|
-
|
|
|
27
|
-
| **GLM 5.3 Flash**
|
|
28
|
-
| **
|
|
29
|
-
| **
|
|
30
|
-
| **
|
|
31
|
-
| **
|
|
32
|
-
| **
|
|
33
|
-
| **
|
|
34
|
-
| **
|
|
35
|
-
| **Gemini 3.
|
|
36
|
-
| **
|
|
37
|
-
| **
|
|
38
|
-
| **
|
|
39
|
-
| **
|
|
40
|
-
| **
|
|
41
|
-
| **
|
|
42
|
-
| **
|
|
43
|
-
| **
|
|
44
|
-
| **
|
|
45
|
-
| **
|
|
46
|
-
| **
|
|
47
|
-
| **
|
|
48
|
-
| **
|
|
49
|
-
| **
|
|
50
|
-
| **
|
|
51
|
-
| **
|
|
52
|
-
| **
|
|
53
|
-
| **
|
|
54
|
-
| **
|
|
55
|
-
| **
|
|
56
|
-
| **
|
|
57
|
-
| **
|
|
58
|
-
| **
|
|
59
|
-
| **
|
|
60
|
-
| **
|
|
61
|
-
| **
|
|
62
|
-
| **GPT-5.
|
|
63
|
-
| **Claude
|
|
64
|
-
| **Claude
|
|
65
|
-
| **
|
|
66
|
-
| **
|
|
29
|
+
| Model | Provider | Cost per intelligence | Intelligence Score | Input (per 1M) | Cached input | Output (per 1M) | Est. session | Context window |
|
|
30
|
+
| -------------------------- | --------- | --------------------- | ------------------ | ----------------------------- | ----------------------------- | ----------------------------- | ------------ | -------------- |
|
|
31
|
+
| **GLM 5.3 Flash** | Z.ai | **~$0.0028** | **46.2** | $0.15 | $0.03 | $0.50 | ~$0.13 | 1M |
|
|
32
|
+
| **GPT-5.6 Luna** | OpenAI | **~$0.0051** | **43.4** | $0.20 | $0.02 | $1.20 | ~$0.22 | 1M |
|
|
33
|
+
| **MiniMax M3** | MiniMax | **~$0.0076** | **35.7** | $0.60 (≤512K) / $1.20 (>512K) | $0.12 (≤512K) / $0.24 (>512K) | $2.40 (≤512K) / $4.80 (>512K) | ~$0.27 | 1M |
|
|
34
|
+
| **DeepSeek V4 Flash 0731** | DeepSeek | **~$0.0086** | **40.8** | $0.44 | $0.014 | $1.32 | ~$0.35 | 1M |
|
|
35
|
+
| **MiMo V2.5 Pro** | Xiaomi | **~$0.0092** | **32.6** | $0.435 | $0.0036 | $0.87 | ~$0.30 | 1M |
|
|
36
|
+
| **MiniMax M3 Priority** | MiniMax | **~$0.0115** | **35.7** | $0.90 (≤512K) / $1.80 (>512K) | $0.18 (≤512K) / $0.36 (>512K) | $3.60 (≤512K) / $7.20 (>512K) | ~$0.41 | 1M |
|
|
37
|
+
| **Gemini 3.8 Flash** | Google | **~$0.0159** | **47.1** | $0.75 | $0.075 | $3.75 | ~$0.75 | 1M |
|
|
38
|
+
| **Gemini 3.7 Flash** | Google | **~$0.0166** | **45.2** | $0.75 | $0.075 | $3.75 | ~$0.75 | 1M |
|
|
39
|
+
| **Gemini 3.6 Flash** | Google | **~$0.0186** | **40.3** | $0.75 | $0.075 | $3.75 | ~$0.75 | 1M |
|
|
40
|
+
| **GLM 5.3** | Z.ai | **~$0.0235** | **48.6** | $1.40 | $0.26 | $4.40 | ~$1.14 | 1M |
|
|
41
|
+
| **DeepSeek V4 Pro 0813** | DeepSeek | **~$0.0252** | **42.1** | $1.32 | $0.044 | $3.96 | ~$1.06 | 1M |
|
|
42
|
+
| **Qwen 3.8 Max (0803)²⁵** | DashScope | **~$0.0300** | **53.4** | $2.00 | $0.25 | $6.00 | ~$1.60 | 1M |
|
|
43
|
+
| **Grok 4.6** | xAI | **~$0.0316** | **50.6** | $2.00 | $0.50 | $6.00 | ~$1.60 | 500K |
|
|
44
|
+
| **Qwen 3.8 Max (0902)²⁶** | DashScope | **~$0.0341** | **46.9** | $2.00 | $0.25 | $6.00 | ~$1.60 | 1M |
|
|
45
|
+
| **Grok 4.5** | xAI | **~$0.0352** | **45.5** | $2.00 | $0.30 | $6.00 | ~$1.60 | 500K |
|
|
46
|
+
| **Claude Sonnet 5** | Anthropic | **~$0.0443** | **45.1** | $2.00 | $0.20 | $10.00 | ~$2.00 | 1M |
|
|
47
|
+
| **GPT-5.6 Terra** | OpenAI | **~$0.0470** | **46.8** | $2.00 | $0.20 | $12.00 | ~$2.20 | 1M |
|
|
48
|
+
| **Kimi K3** | Moonshot | **~$0.0598** | **50.2** | $3.00 | $0.30 | $15.00 | ~$3.00 | 1M |
|
|
49
|
+
| **Gemini 3.1 Pro** | Google | **~$0.0599** | **36.7** | $2.00¹ | $0.20 | $12.00¹ | ~$2.20 | 1M |
|
|
50
|
+
| **GPT-5.6 Sol** | OpenAI | **~$0.0780** | **51.3** | $4.00 | $0.40 | $20.00 | ~$4.00 | 1M |
|
|
51
|
+
| **Claude Opus 5** | Anthropic | **~$0.0924** | **54.1** | $5.00 | $0.50 | $25.00 | ~$5.00 | 1M |
|
|
52
|
+
| **Claude Fable 5.1** | Anthropic | **~$0.1761** | **56.8** | $10.00 | $0.25 | $50.00 | ~$10.00 | 1M |
|
|
53
|
+
| **GPT-6 Astra** | OpenAI | **~$0.1828** | **54.7** | $10.00 | $1.00 | $50.00 | ~$10.00 | 1M |
|
|
54
|
+
| **Claude Fable 5** | Anthropic | **~$0.1880** | **53.2** | $10.00 | $1.00 | $50.00 | ~$10.00 | 1M |
|
|
55
|
+
| **MiMo V2.5** | Xiaomi | — | — | $0.14 | $0.0028 | $0.28 | ~$0.10 | 1M |
|
|
56
|
+
| **Qwen 3.7 Plus** | DashScope | — | — | $0.40 (≤256K) / $1.20 (>256K) | $0.04 | $1.60 (≤256K) / $4.80 (>256K) | ~$0.36 | 1M |
|
|
57
|
+
| **Qwen 3.6 Plus** | DashScope | — | — | $0.50 | $0.05 | $3.00 | ~$0.55 | 1M |
|
|
58
|
+
| **GPT-5.4 mini** | OpenAI | — | — | $0.75 | $0.075 | $4.50 | ~$0.83 | 400K |
|
|
59
|
+
| **Kimi K2.6** | Moonshot | — | — | $0.95 | $0.16 | $4.00 | ~$0.88 | 256K |
|
|
60
|
+
| **Kimi K2.7 Code** | Moonshot | — | — | $0.95 | $0.19 | $4.00 | ~$0.88 | 262K |
|
|
61
|
+
| **GLM 5V Turbo** | Z.ai | — | — | $1.20 | $0.24 | $4.00 | ~$1.00 | 200K |
|
|
62
|
+
| **GLM 5.1** | Z.ai | — | — | $1.40 | $0.26 | $4.40 | ~$1.14 | 200K |
|
|
63
|
+
| **GLM 5.2** | Z.ai | — | — | $1.40 | $0.26 | $4.40 | ~$1.14 | 1M |
|
|
64
|
+
| **Gemini 3.5 Flash** | Google | — | — | $1.50 | $0.15 | $9.00 | ~$1.65 | 1M |
|
|
65
|
+
| **Qwen 3.7 Max** | DashScope | — | — | $2.50 (≤1M) | $0.25 | $7.50 (≤1M) | ~$2.00 | 1M |
|
|
66
|
+
| **GPT-5.4** | OpenAI | — | — | $2.50 | $0.25 | $15.00 | ~$2.75 | 1M |
|
|
67
|
+
| **Claude Sonnet 4.6** | Anthropic | — | — | $3.00 | $0.30 | $15.00 | ~$3.00 | 1M |
|
|
68
|
+
| **Claude Opus 4.8** | Anthropic | — | — | $5.00 | $0.50 | $25.00 | ~$5.00 | 1M |
|
|
69
|
+
| **Claude Opus 4.7** | Anthropic | — | — | $5.00 | $0.50 | $25.00 | ~$5.00 | 1M |
|
|
70
|
+
| **GPT-5.5** | OpenAI | — | — | $5.00 | $0.50 | $30.00 | ~$5.50 | 1M |
|
|
71
|
+
| **Gemini 3.8 Flash Cyber** | Google | — | — | — | — | — | — | — |
|
|
72
|
+
|
|
73
|
+
> **Historical footnotes:** The numbered notes below preserve release and pricing context from the prior September 5 snapshot. They are not the current AA ranking source; use the table above and the September 6 OpenRouter refresh note for current scores.
|
|
67
74
|
|
|
68
75
|
¹ Gemini 3.1 Pro pricing applies to prompts ≤200K tokens.
|
|
69
76
|
|
|
@@ -83,7 +90,7 @@ All models are listed together below, sorted by Cost per intelligence ascending
|
|
|
83
90
|
|
|
84
91
|
¹⁰ **Claude Opus 5** was released on **July 24, 2026**. AA Intelligence Index score (**63.1**) confirmed by [Artificial Analysis](https://artificialanalysis.ai/models/claude-opus-5). Pricing is $5.00 / $0.50 / $25.00 per MTok input/cached/output (same as Opus 4.8). 1M context window, text + image input, adaptive reasoning. Also available in Fast mode ($10/$50 per MTok input/output) and Batch ($2.50/$12.50 per MTok input/output). Uses the newer Claude tokenizer (~30% more tokens than pre-4.7 models). See [Anthropic's Opus 5 announcement](https://www.anthropic.com/news/claude-opus-5).
|
|
85
92
|
|
|
86
|
-
¹¹ **Qwen 3.8 Max
|
|
93
|
+
¹¹ **Historical Qwen 3.8 Max launch record:** the August 3, 2026 launch package used the unversioned `qwen3.8-max` name. An earlier Artificial Analysis snapshot reported **58.1**; the current OpenRouter ranking now distinguishes the unversioned **0803** row (**53.4**) from the separate **0902** row (**46.9**). Use the current rows and footnotes ²⁵–²⁶ for comparisons rather than carrying the earlier score forward. See the [Qwen Cloud model page](https://www.qwencloud.com/models/qwen3.8-max), [AA model page](https://artificialanalysis.ai/models/qwen3-8-max), and [OpenRouter model page](https://openrouter.ai/qwen/qwen3.8-max).
|
|
87
94
|
|
|
88
95
|
¹² **Grok 4.5** (xAI, released **July 8, 2026**) is now a **GitHub Copilot native** model (GA, Versatile). AA Intelligence Index **55.8** (high), Coding **72.4**, Agentic **48.9**. 500K context window, text + image input. Priced at $2.00 / $0.50 / $6.00 per 1M input/cached/output (75% cache discount; Copilot long-context >200K tier $4.00 / $1.00 / $12.00). See the [GitHub Copilot models & pricing page](https://docs.github.com/en/copilot/reference/copilot-billing/models-and-pricing).
|
|
89
96
|
|
|
@@ -111,6 +118,10 @@ All models are listed together below, sorted by Cost per intelligence ascending
|
|
|
111
118
|
|
|
112
119
|
²⁴ **GPT-6 Astra** was released on **September 3, 2026** as `gpt-6-astra`, an OpenAI API model and **GitHub Copilot native** model (GA, Powerful). It supports reasoning, text + image input, text output, and a 1M-token context window. Standard pricing is **$10.00 / $1.00 / $50.00** per 1M input/cached/output tokens; cache writes are $12.50/M, and prompts above 272K input tokens use long-context rates. The modeled session cost is ~$10.00 and CPI is ~$0.163. See the [GPT-6 Astra announcement](https://openai.com/index/gpt-6-astra/), [OpenAI API pricing](https://developers.openai.com/api/docs/pricing), and [GitHub Copilot models & pricing](https://docs.github.com/en/copilot/reference/copilot-billing/models-and-pricing).
|
|
113
120
|
|
|
121
|
+
²⁵ **Qwen 3.8 Max (0803)** is the August 3, 2026 launch checkpoint exposed by OpenRouter under the unversioned slug [`qwen/qwen3.8-max`](https://openrouter.ai/qwen/qwen3.8-max). The September 7 ranking snapshot gives it an AA Intelligence Index of **53.4**. It shares Qwen Cloud's **$2.00 / $0.25 / $6.00** input/cached/output rates, 1M context, 131K maximum output, and text/image/video input with the 0902 snapshot. OpenRouter describes it as superseded by 0902; the validated DashScope setup continues to use provider ID `qwen3.8-max` and the proxy forwards that ID unchanged.
|
|
122
|
+
|
|
123
|
+
²⁶ **Qwen 3.8 Max (0902)** is a separate OpenRouter snapshot, [`qwen/qwen3.8-max-0902`](https://openrouter.ai/qwen/qwen3.8-max-0902), released September 4, 2026. The September 7 ranking snapshot gives it an AA Intelligence Index of **46.9**. It has the same published **$2.00 / $0.25 / $6.00** input/cached/output rates, 1M context, 131K maximum output, and text/image/video input. The OpenRouter slug is not automatically a valid DashScope model ID; see [the Qwen setup guide](models/qwen.md) before changing the upstream URL.
|
|
124
|
+
|
|
114
125
|
Cost per intelligence = estimated session cost ÷ Intelligence Index score. Session cost assumes ~10K input + ~2K output tokens per turn, 50 turns.
|
|
115
126
|
|
|
116
127
|
> **Notes:**
|
|
@@ -131,7 +142,7 @@ Cost per intelligence = estimated session cost ÷ Intelligence Index score. Sess
|
|
|
131
142
|
|
|
132
143
|
> **How long does 7,000 credits last?** A Pro+ subscriber running 50-turn sessions could afford roughly **13 GPT-5.5 sessions** or **23 Opus sessions** per month — or mix and match. (Multiply session cost by 100 to convert to AI credits.)
|
|
133
144
|
|
|
134
|
-
> Prices last verified: September
|
|
145
|
+
> Prices last verified: September 7, 2026. Always check the official pages for the latest rates:
|
|
135
146
|
>
|
|
136
147
|
> - [GitHub Copilot models & pricing](https://docs.github.com/en/copilot/reference/copilot-billing/models-and-pricing)
|
|
137
148
|
> - [OpenAI pricing](https://openai.com/api/pricing/)
|
|
@@ -144,3 +155,5 @@ Cost per intelligence = estimated session cost ÷ Intelligence Index score. Sess
|
|
|
144
155
|
> - [MiMo pricing](https://mimo.mi.com/docs/en-US/pricing)
|
|
145
156
|
> - [MiniMax pricing](https://platform.minimax.io/docs/pricing/overview)
|
|
146
157
|
> - [Z.ai (GLM) pricing](https://docs.z.ai/guides/overview/pricing)
|
|
158
|
+
> - [Kimi pricing](https://platform.kimi.ai/docs/pricing)
|
|
159
|
+
> - [xAI model pricing](https://docs.x.ai/docs/models)
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "copilot-custom-endpoint",
|
|
3
|
-
"version": "
|
|
3
|
+
"version": "3.0.0",
|
|
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
|
+
}
|
package/proxy/qwen-proxy.mjs
CHANGED
|
@@ -6,6 +6,9 @@ import { createProxy } from '../lib/create-proxy.mjs'
|
|
|
6
6
|
/**
|
|
7
7
|
* Supported model scope for this proxy:
|
|
8
8
|
* - Validated with `qwen3.8-max`, `qwen3.7-plus`, and `qwen3.7-max`.
|
|
9
|
+
* - The rewrite can also be used with OpenRouter model IDs such as
|
|
10
|
+
* `qwen/qwen3.8-max-0902`; the proxy forwards model IDs unchanged. That route
|
|
11
|
+
* is not part of the validated DashScope setup.
|
|
9
12
|
* - Expected to work for any Qwen3 hybrid-thinking model (qwen3-* series)
|
|
10
13
|
* that supports the `enable_thinking` top-level field on DashScope's
|
|
11
14
|
* OpenAI-compatible surface.
|
|
@@ -33,9 +36,13 @@ Starts a local HTTP proxy that conditionally injects enable_thinking: false
|
|
|
33
36
|
when the request includes a tools array, letting Qwen hybrid-thinking models
|
|
34
37
|
show reasoning in plain chat while keeping tool loops stable.
|
|
35
38
|
|
|
39
|
+
The proxy forwards the requested model ID unchanged, including the DashScope
|
|
40
|
+
ID qwen3.8-max and OpenRouter ID qwen/qwen3.8-max-0902.
|
|
41
|
+
|
|
36
42
|
Environment variables:
|
|
37
43
|
QWEN_PROXY_PORT Local listen port. Default: 3458 (falls back to PORT)
|
|
38
|
-
QWEN_UPSTREAM_URL Upstream
|
|
44
|
+
QWEN_UPSTREAM_URL Upstream chat-completions URL (DashScope by default;
|
|
45
|
+
OpenRouter can be selected for snapshot IDs).
|
|
39
46
|
Default: https://dashscope-intl.aliyuncs.com/compatible-mode/v1/chat/completions
|
|
40
47
|
QWEN_PROXY_DISABLE_THINKING_WITH_TOOLS
|
|
41
48
|
Inject enable_thinking: false when tools are present.
|