Skip to content

Commit a3bd259

Browse files
gustavodiasdevgustavodiasdev
committed
feat: add general-purpose Qwen AI models and bump to v1.2.0
Expande modelos disponíveis além dos Coder para incluir qwen3-max (flagship), qwen-plus-latest (balanceado com thinking mode), qwen3-235b-a22b (maior MoE open-weight) e qwen-flash (ultra-rápido). Mudanças principais: - Adicionados 4 modelos gerais: qwen3-max, qwen-plus-latest, qwen3-235b-a22b, qwen-flash - Campo reasoning agora é dinâmico por modelo (true para plus e 235b) - READMEs atualizados com tabelas separadas por categoria - Versão atualizada para 1.2.0
1 parent 4f68548 commit a3bd259

6 files changed

Lines changed: 87 additions & 10 deletions

File tree

‎README.md‎

Lines changed: 16 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -8,7 +8,7 @@
88
<img src="assets/screenshot.png" alt="OpenCode with Qwen Code" width="800">
99
</p>
1010

11-
**Authenticate OpenCode CLI with your qwen.ai account.** This plugin enables you to use Qwen3-Coder models with **2,000 free requests per day** - no API key or credit card required!
11+
**Authenticate OpenCode CLI with your qwen.ai account.** This plugin enables you to use Qwen models (Coder, Max, Plus and more) with **2,000 free requests per day** - no API key or credit card required!
1212

1313
[🇧🇷 Leia em Português](./README.pt-BR.md)
1414

@@ -69,22 +69,35 @@ Select **"Qwen Code (qwen.ai OAuth)"**
6969
7070
## 🎯 Available Models
7171

72+
### Coding Models
73+
7274
| Model | Context | Max Output | Best For |
7375
|-------|---------|------------|----------|
7476
| `qwen3-coder-plus` | 1M tokens | 64K tokens | Complex coding tasks |
75-
| `qwen3-coder-flash` | 1M tokens | 64K tokens | Fast responses |
77+
| `qwen3-coder-flash` | 1M tokens | 64K tokens | Fast coding responses |
78+
79+
### General Purpose Models
80+
81+
| Model | Context | Max Output | Reasoning | Best For |
82+
|-------|---------|------------|-----------|----------|
83+
| `qwen3-max` | 256K tokens | 64K tokens | No | Flagship model, complex reasoning and tool use |
84+
| `qwen-plus-latest` | 128K tokens | 16K tokens | Yes | Balanced quality-speed with thinking mode |
85+
| `qwen3-235b-a22b` | 128K tokens | 32K tokens | Yes | Largest open-weight MoE with thinking mode |
86+
| `qwen-flash` | 1M tokens | 8K tokens | No | Ultra-fast, low-cost simple tasks |
7687

7788
### Using a specific model
7889

7990
```bash
8091
opencode --provider qwen-code --model qwen3-coder-plus
92+
opencode --provider qwen-code --model qwen3-max
93+
opencode --provider qwen-code --model qwen-plus-latest
8194
```
8295

8396
## ⚙️ How It Works
8497

8598
```
8699
┌─────────────────┐ ┌──────────────────┐ ┌─────────────────┐
87-
│ OpenCode CLI │────▶│ qwen.ai OAuth │────▶│ Qwen3-Coder │
100+
│ OpenCode CLI │────▶│ qwen.ai OAuth │────▶│ Qwen Models │
88101
│ │◀────│ (Device Flow) │◀────│ API │
89102
└─────────────────┘ └──────────────────┘ └─────────────────┘
90103
```

‎README.pt-BR.md‎

Lines changed: 16 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -8,7 +8,7 @@
88
<img src="assets/screenshot.png" alt="OpenCode com Qwen Code" width="800">
99
</p>
1010

11-
**Autentique o OpenCode CLI com sua conta qwen.ai.** Este plugin permite usar modelos Qwen3-Coder com **2.000 requisições gratuitas por dia** - sem API key ou cartão de crédito!
11+
**Autentique o OpenCode CLI com sua conta qwen.ai.** Este plugin permite usar modelos Qwen (Coder, Max, Plus e mais) com **2.000 requisições gratuitas por dia** - sem API key ou cartão de crédito!
1212

1313
[🇺🇸 Read in English](./README.md)
1414

@@ -69,22 +69,35 @@ Selecione **"Qwen Code (qwen.ai OAuth)"**
6969
7070
## 🎯 Modelos Disponíveis
7171

72+
### Modelos de Código
73+
7274
| Modelo | Contexto | Max Output | Melhor Para |
7375
|--------|----------|------------|-------------|
7476
| `qwen3-coder-plus` | 1M tokens | 64K tokens | Tarefas complexas de código |
75-
| `qwen3-coder-flash` | 1M tokens | 64K tokens | Respostas rápidas |
77+
| `qwen3-coder-flash` | 1M tokens | 64K tokens | Respostas rápidas de código |
78+
79+
### Modelos de Propósito Geral
80+
81+
| Modelo | Contexto | Max Output | Reasoning | Melhor Para |
82+
|--------|----------|------------|-----------|-------------|
83+
| `qwen3-max` | 256K tokens | 64K tokens | Não | Modelo flagship, raciocínio complexo e tool use |
84+
| `qwen-plus-latest` | 128K tokens | 16K tokens | Sim | Equilíbrio qualidade-velocidade com thinking mode |
85+
| `qwen3-235b-a22b` | 128K tokens | 32K tokens | Sim | Maior modelo open-weight MoE com thinking mode |
86+
| `qwen-flash` | 1M tokens | 8K tokens | Não | Ultra-rápido, baixo custo para tarefas simples |
7687

7788
### Usando um modelo específico
7889

7990
```bash
8091
opencode --provider qwen-code --model qwen3-coder-plus
92+
opencode --provider qwen-code --model qwen3-max
93+
opencode --provider qwen-code --model qwen-plus-latest
8194
```
8295

8396
## ⚙️ Como Funciona
8497

8598
```
8699
┌─────────────────┐ ┌──────────────────┐ ┌─────────────────┐
87-
│ OpenCode CLI │────▶│ qwen.ai OAuth │────▶│ Qwen3-Coder │
100+
│ OpenCode CLI │────▶│ qwen.ai OAuth │────▶│ Qwen Models │
88101
│ │◀────│ (Device Flow) │◀────│ API │
89102
└─────────────────┘ └──────────────────┘ └─────────────────┘
90103
```

‎bun.lock‎

Lines changed: 8 additions & 0 deletions
Some generated files are not rendered by default. Learn more about customizing how changed files appear on GitHub.

‎package.json‎

Lines changed: 5 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,7 +1,7 @@
11
{
22
"name": "opencode-qwencode-auth",
3-
"version": "1.1.0",
4-
"description": "Qwen OAuth authentication plugin for OpenCode - Access Qwen3-Coder models with your qwen.ai account",
3+
"version": "1.2.0",
4+
"description": "Qwen OAuth authentication plugin for OpenCode - Access Qwen AI models (Coder, Max, Plus and more) with your qwen.ai account",
55
"module": "index.ts",
66
"type": "module",
77
"scripts": {
@@ -14,6 +14,9 @@
1414
"qwen",
1515
"qwen-code",
1616
"qwen3-coder",
17+
"qwen3-max",
18+
"qwen-plus",
19+
"qwen-flash",
1720
"oauth",
1821
"authentication",
1922
"ai",

‎src/constants.ts‎

Lines changed: 41 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -35,14 +35,16 @@ export const QWEN_API_CONFIG = {
3535
export const CALLBACK_PORT = 14561;
3636

3737
// Available Qwen models through OAuth
38-
// Baseado nos modelos disponíveis no qwen-code
38+
// Baseado nos modelos disponíveis no qwen-code + modelos gerais via portal.qwen.ai
3939
export const QWEN_MODELS = {
40+
// --- Coding Models ---
4041
'qwen3-coder-plus': {
4142
id: 'qwen3-coder-plus',
4243
name: 'Qwen3 Coder Plus',
4344
contextWindow: 1048576, // 1M tokens
4445
maxOutput: 65536, // 64K tokens
4546
description: 'Most capable Qwen coding model with 1M context window',
47+
reasoning: false,
4648
cost: { input: 0, output: 0 }, // Free via OAuth
4749
},
4850
'qwen3-coder-flash': {
@@ -51,6 +53,44 @@ export const QWEN_MODELS = {
5153
contextWindow: 1048576,
5254
maxOutput: 65536,
5355
description: 'Faster Qwen coding model for quick responses',
56+
reasoning: false,
57+
cost: { input: 0, output: 0 },
58+
},
59+
// --- General Purpose Models ---
60+
'qwen3-max': {
61+
id: 'qwen3-max',
62+
name: 'Qwen3 Max',
63+
contextWindow: 262144, // 256K tokens
64+
maxOutput: 65536, // 64K tokens
65+
description: 'Flagship ~1T parameter MoE model, best for complex reasoning and tool use',
66+
reasoning: false,
67+
cost: { input: 0, output: 0 },
68+
},
69+
'qwen-plus-latest': {
70+
id: 'qwen-plus-latest',
71+
name: 'Qwen Plus',
72+
contextWindow: 131072, // 128K tokens
73+
maxOutput: 16384, // 16K tokens
74+
description: 'Balanced model with thinking mode, good quality-speed tradeoff',
75+
reasoning: true,
76+
cost: { input: 0, output: 0 },
77+
},
78+
'qwen3-235b-a22b': {
79+
id: 'qwen3-235b-a22b',
80+
name: 'Qwen3 235B-A22B',
81+
contextWindow: 131072, // 128K tokens
82+
maxOutput: 32768, // 32K tokens
83+
description: 'Largest open-weight Qwen3 MoE model with thinking mode',
84+
reasoning: true,
85+
cost: { input: 0, output: 0 },
86+
},
87+
'qwen-flash': {
88+
id: 'qwen-flash',
89+
name: 'Qwen Flash',
90+
contextWindow: 1048576, // 1M tokens
91+
maxOutput: 8192, // 8K tokens
92+
description: 'Ultra-fast and low-cost model for simple tasks',
93+
reasoning: false,
5494
cost: { input: 0, output: 0 },
5595
},
5696
} as const;

‎src/index.ts‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -214,7 +214,7 @@ export const QwenAuthPlugin = async (_input: unknown) => {
214214
{
215215
id: m.id,
216216
name: m.name,
217-
reasoning: false,
217+
reasoning: m.reasoning,
218218
limit: { context: m.contextWindow, output: m.maxOutput },
219219
cost: m.cost,
220220
modalities: { input: ['text'], output: ['text'] },

0 commit comments

Comments
 (0)