Skip to content

Commit 7fef09a

Browse files
warren618warren618
committed
chore(providers): refresh default model ids
1 parent 33105a5 commit 7fef09a

16 files changed

Lines changed: 114 additions & 62 deletions

‎README.md‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -452,8 +452,8 @@ Vibe-Trading is a tool-heavy agent — skills, backtests, memory, and swarms all
452452

453453
| Tier | Examples | When to use |
454454
|------|----------|-------------|
455-
| **Best** | `anthropic/claude-opus-4.7`, `anthropic/claude-sonnet-4.6`, `openai/gpt-5.4`, `google/gemini-3.1-pro-preview` | Complex swarms (3+ agents), long research sessions, paper-grade analysis |
456-
| **Sweet spot** (default) | `deepseek-v4-pro`, `deepseek/deepseek-v4-pro`, `x-ai/grok-4.20`, `z-ai/glm-5.1`, `moonshotai/kimi-k2.5`, `qwen/qwen3-max-thinking` | Daily driver — reliable tool-calling at ~1/10 the cost |
455+
| **Best** | `anthropic/claude-opus-4.7`, `anthropic/claude-sonnet-4.6`, `openai/gpt-5.5-pro`, `google/gemini-3.5-flash` | Complex swarms (3+ agents), long research sessions, paper-grade analysis |
456+
| **Sweet spot** (default) | `deepseek-v4-pro`, `deepseek/deepseek-v4-pro`, `x-ai/grok-4.20`, `z-ai/glm-5.1`, `moonshotai/kimi-k2.6`, `qwen/qwen3-max-thinking` | Daily driver — reliable tool-calling at ~1/10 the cost |
457457
| **Avoid for agent use** | `*-nano`, `*-flash-lite`, `*-coder-next`, small / distilled variants | Tool-calling is unreliable — the agent will appear to "answer from memory" instead of loading skills or running backtests |
458458

459459
The default `agent/.env.example` ships with DeepSeek official API + `deepseek-v4-pro`; OpenRouter users can use `deepseek/deepseek-v4-pro`.

‎README_ar.md‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -453,8 +453,8 @@ Vibe-Trading وكيل كثيف الأدوات؛ المهارات والاختب
453453

454454
| المستوى | أمثلة | متى تستخدمه |
455455
|------|----------|-------------|
456-
| **الأفضل** | `anthropic/claude-opus-4.7`, `anthropic/claude-sonnet-4.6`, `openai/gpt-5.4`, `google/gemini-3.1-pro-preview` | أسراب معقدة (3+ وكلاء)، جلسات بحث طويلة، تحليل بمستوى ورقة علمية |
457-
| **النقطة المثلى** (افتراضي) | `deepseek-v4-pro`, `deepseek/deepseek-v4-pro`, `x-ai/grok-4.20`, `z-ai/glm-5.1`, `moonshotai/kimi-k2.5`, `qwen/qwen3-max-thinking` | الاستخدام اليومي، tool-calling موثوق بنحو عُشر التكلفة |
456+
| **الأفضل** | `anthropic/claude-opus-4.7`, `anthropic/claude-sonnet-4.6`, `openai/gpt-5.5-pro`, `google/gemini-3.5-flash` | أسراب معقدة (3+ وكلاء)، جلسات بحث طويلة، تحليل بمستوى ورقة علمية |
457+
| **النقطة المثلى** (افتراضي) | `deepseek-v4-pro`, `deepseek/deepseek-v4-pro`, `x-ai/grok-4.20`, `z-ai/glm-5.1`, `moonshotai/kimi-k2.6`, `qwen/qwen3-max-thinking` | الاستخدام اليومي، tool-calling موثوق بنحو عُشر التكلفة |
458458
| **تجنبها لاستخدام الوكيل** | `*-nano`, `*-flash-lite`, `*-coder-next`, small / distilled variants | tool-calling غير موثوق؛ سيبدو الوكيل وكأنه "يجيب من الذاكرة" بدلاً من تحميل المهارات أو تشغيل الاختبارات الرجعية |
459459

460460
يأتي `agent/.env.example` افتراضياً مع DeepSeek official API + `deepseek-v4-pro`; ويمكن لمستخدمي OpenRouter استخدام `deepseek/deepseek-v4-pro`.

‎README_ja.md‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -453,8 +453,8 @@ Vibe-Trading は tool-heavy agent です。skills、backtests、memory、swarms
453453

454454
| Tier | Examples | When to use |
455455
|------|----------|-------------|
456-
| **Best** | `anthropic/claude-opus-4.7`, `anthropic/claude-sonnet-4.6`, `openai/gpt-5.4`, `google/gemini-3.1-pro-preview` | 複雑な swarms（3+ agents）、長い research sessions、paper-grade analysis |
457-
| **Sweet spot** (default) | `deepseek-v4-pro`, `deepseek/deepseek-v4-pro`, `x-ai/grok-4.20`, `z-ai/glm-5.1`, `moonshotai/kimi-k2.5`, `qwen/qwen3-max-thinking` | 日常使い。信頼できる tool-calling を約 1/10 の cost で |
456+
| **Best** | `anthropic/claude-opus-4.7`, `anthropic/claude-sonnet-4.6`, `openai/gpt-5.5-pro`, `google/gemini-3.5-flash` | 複雑な swarms（3+ agents）、長い research sessions、paper-grade analysis |
457+
| **Sweet spot** (default) | `deepseek-v4-pro`, `deepseek/deepseek-v4-pro`, `x-ai/grok-4.20`, `z-ai/glm-5.1`, `moonshotai/kimi-k2.6`, `qwen/qwen3-max-thinking` | 日常使い。信頼できる tool-calling を約 1/10 の cost で |
458458
| **Avoid for agent use** | `*-nano`, `*-flash-lite`, `*-coder-next`, small / distilled variants | Tool-calling が不安定です。agent は skills 読み込みや backtests 実行ではなく「記憶から答えている」ように見えます |
459459

460460
既定の `agent/.env.example` は DeepSeek official API + `deepseek-v4-pro` で出荷されています。OpenRouter users は `deepseek/deepseek-v4-pro` を利用できます。

‎README_ko.md‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -453,8 +453,8 @@ Vibe-Trading은 tool-heavy agent입니다. skills, backtests, memory, swarms가
453453

454454
| 등급 | 예시 | 사용 시점 |
455455
|------|------|-----------|
456-
| **Best** | `anthropic/claude-opus-4.7`, `anthropic/claude-sonnet-4.6`, `openai/gpt-5.4`, `google/gemini-3.1-pro-preview` | 복잡한 swarms(3+ agents), 긴 리서치 세션, 논문급 분석 |
457-
| **Sweet spot**(기본값) | `deepseek-v4-pro`, `deepseek/deepseek-v4-pro`, `x-ai/grok-4.20`, `z-ai/glm-5.1`, `moonshotai/kimi-k2.5`, `qwen/qwen3-max-thinking` | Daily driver — 약 1/10 비용으로 안정적인 tool-calling |
456+
| **Best** | `anthropic/claude-opus-4.7`, `anthropic/claude-sonnet-4.6`, `openai/gpt-5.5-pro`, `google/gemini-3.5-flash` | 복잡한 swarms(3+ agents), 긴 리서치 세션, 논문급 분석 |
457+
| **Sweet spot**(기본값) | `deepseek-v4-pro`, `deepseek/deepseek-v4-pro`, `x-ai/grok-4.20`, `z-ai/glm-5.1`, `moonshotai/kimi-k2.6`, `qwen/qwen3-max-thinking` | Daily driver — 약 1/10 비용으로 안정적인 tool-calling |
458458
| **Agent 사용 시 피할 것** | `*-nano`, `*-flash-lite`, `*-coder-next`, small / distilled variants | tool-calling이 불안정합니다. agent가 skills를 로드하거나 backtest를 실행하는 대신 "기억에서 답하는" 것처럼 보일 수 있습니다. |
459459

460460
기본 `agent/.env.example`은 DeepSeek official API + `deepseek-v4-pro`를 포함합니다. OpenRouter 사용자는 `deepseek/deepseek-v4-pro`를 사용할 수 있습니다.

‎README_zh.md‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -452,8 +452,8 @@ Vibe-Trading 是高度依赖工具的智能体：skills、backtests、memory 和
452452

453453
| 档位 | 示例 | 使用场景 |
454454
|------|------|----------|
455-
| **Best** | `anthropic/claude-opus-4.7`, `anthropic/claude-sonnet-4.6`, `openai/gpt-5.4`, `google/gemini-3.1-pro-preview` | 复杂 swarms（3+ agents）、长研究 sessions、论文级分析 |
456-
| **Sweet spot**（默认） | `deepseek-v4-pro`, `deepseek/deepseek-v4-pro`, `x-ai/grok-4.20`, `z-ai/glm-5.1`, `moonshotai/kimi-k2.5`, `qwen/qwen3-max-thinking` | 日常主力，约 1/10 成本下具备可靠工具调用 |
455+
| **Best** | `anthropic/claude-opus-4.7`, `anthropic/claude-sonnet-4.6`, `openai/gpt-5.5-pro`, `google/gemini-3.5-flash` | 复杂 swarms（3+ agents）、长研究 sessions、论文级分析 |
456+
| **Sweet spot**（默认） | `deepseek-v4-pro`, `deepseek/deepseek-v4-pro`, `x-ai/grok-4.20`, `z-ai/glm-5.1`, `moonshotai/kimi-k2.6`, `qwen/qwen3-max-thinking` | 日常主力，约 1/10 成本下具备可靠工具调用 |
457457
| **避免用于 agent** | `*-nano`, `*-flash-lite`, `*-coder-next`, 小型 / 蒸馏变体 | 工具调用不可靠，智能体会看起来像是在“凭记忆回答”，而不是加载 skills 或运行回测 |
458458

459459
默认 `agent/.env.example` 使用 DeepSeek 官方 API + `deepseek-v4-pro`；OpenRouter 用户可以使用 `deepseek/deepseek-v4-pro`。

‎agent/.env.example‎

Lines changed: 8 additions & 8 deletions
Original file line numberDiff line numberDiff line change
@@ -4,13 +4,13 @@
44

55
# --- OpenRouter (recommended, multi-model gateway) ---
66
LANGCHAIN_PROVIDER=openrouter
7-
LANGCHAIN_MODEL_NAME=deepseek/deepseek-v3.2
7+
LANGCHAIN_MODEL_NAME=deepseek/deepseek-v4-pro
88
OPENROUTER_API_KEY=sk-or-v1-your-key-here
99
OPENROUTER_BASE_URL=https://openrouter.ai/api/v1
1010

1111
# --- OpenAI ---
1212
# LANGCHAIN_PROVIDER=openai
13-
# LANGCHAIN_MODEL_NAME=gpt-4o
13+
# LANGCHAIN_MODEL_NAME=gpt-5.5-instant
1414
# OPENAI_API_KEY=sk-xxx
1515
# OPENAI_BASE_URL=https://api.openai.com/v1
1616

@@ -23,37 +23,37 @@ OPENROUTER_BASE_URL=https://openrouter.ai/api/v1
2323

2424
# --- DeepSeek ---
2525
# LANGCHAIN_PROVIDER=deepseek
26-
# LANGCHAIN_MODEL_NAME=deepseek-chat
26+
# LANGCHAIN_MODEL_NAME=deepseek-v4-pro
2727
# DEEPSEEK_API_KEY=sk-xxx
2828
# DEEPSEEK_BASE_URL=https://api.deepseek.com/v1
2929

3030
# --- Gemini ---
3131
# LANGCHAIN_PROVIDER=gemini
32-
# LANGCHAIN_MODEL_NAME=gemini-2.5-flash
32+
# LANGCHAIN_MODEL_NAME=gemini-3.5-flash
3333
# GEMINI_API_KEY=xxx
3434
# GEMINI_BASE_URL=https://generativelanguage.googleapis.com/v1beta/openai/
3535

3636
# --- Groq ---
3737
# LANGCHAIN_PROVIDER=groq
38-
# LANGCHAIN_MODEL_NAME=llama-3.3-70b-versatile
38+
# LANGCHAIN_MODEL_NAME=meta-llama/llama-4-maverick-17b-128e-instruct
3939
# GROQ_API_KEY=gsk_xxx
4040
# GROQ_BASE_URL=https://api.groq.com/openai/v1
4141

4242
# --- DashScope / Qwen ---
4343
# LANGCHAIN_PROVIDER=dashscope
44-
# LANGCHAIN_MODEL_NAME=qwen-plus
44+
# LANGCHAIN_MODEL_NAME=qwen-plus-latest
4545
# DASHSCOPE_API_KEY=sk-xxx
4646
# DASHSCOPE_BASE_URL=https://dashscope.aliyuncs.com/compatible-mode/v1
4747

4848
# --- Zhipu ---
4949
# LANGCHAIN_PROVIDER=zhipu
50-
# LANGCHAIN_MODEL_NAME=glm-4-plus
50+
# LANGCHAIN_MODEL_NAME=glm-5.1
5151
# ZHIPU_API_KEY=xxx
5252
# ZHIPU_BASE_URL=https://open.bigmodel.cn/api/paas/v4
5353

5454
# --- Moonshot / Kimi ---
5555
# LANGCHAIN_PROVIDER=moonshot
56-
# LANGCHAIN_MODEL_NAME=kimi-k2.5
56+
# LANGCHAIN_MODEL_NAME=kimi-k2.6
5757
# MOONSHOT_API_KEY=sk-xxx
5858
# MOONSHOT_BASE_URL=https://api.moonshot.ai/v1
5959

‎agent/SKILL.md‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -14,7 +14,7 @@ env:
1414
description: "OpenAI-compatible API key — only needed for run_swarm (multi-agent teams). All other tools work without it."
1515
required: false
1616
- name: LANGCHAIN_MODEL_NAME
17-
description: "LLM model name for run_swarm (e.g. deepseek/deepseek-v3.2). Only needed if using run_swarm."
17+
description: "LLM model name for run_swarm (e.g. deepseek/deepseek-v4-pro). Only needed if using run_swarm."
1818
required: false
1919
mcp:
2020
command: vibe-trading-mcp

‎agent/cli/_legacy.py‎

Lines changed: 9 additions & 9 deletions
Original file line numberDiff line numberDiff line change
@@ -2695,7 +2695,7 @@ def _handle_prompt_command(
26952695
"key_env": "OPENROUTER_API_KEY",
26962696
"base_env": "OPENROUTER_BASE_URL",
26972697
"base_url": "https://openrouter.ai/api/v1",
2698-
"model": "deepseek/deepseek-v3.2",
2698+
"model": "deepseek/deepseek-v4-pro",
26992699
"key_prefix": "sk-or-",
27002700
"key_placeholder": "sk-or-v1-...",
27012701
},
@@ -2705,7 +2705,7 @@ def _handle_prompt_command(
27052705
"key_env": "DEEPSEEK_API_KEY",
27062706
"base_env": "DEEPSEEK_BASE_URL",
27072707
"base_url": "https://api.deepseek.com/v1",
2708-
"model": "deepseek-chat",
2708+
"model": "deepseek-v4-pro",
27092709
"key_prefix": "sk-",
27102710
"key_placeholder": "sk-...",
27112711
},
@@ -2715,7 +2715,7 @@ def _handle_prompt_command(
27152715
"key_env": "OPENAI_API_KEY",
27162716
"base_env": "OPENAI_BASE_URL",
27172717
"base_url": "https://api.openai.com/v1",
2718-
"model": "gpt-4o",
2718+
"model": "gpt-5.5-instant",
27192719
"key_prefix": "sk-",
27202720
"key_placeholder": "sk-...",
27212721
},
@@ -2725,7 +2725,7 @@ def _handle_prompt_command(
27252725
"key_env": "GEMINI_API_KEY",
27262726
"base_env": "GEMINI_BASE_URL",
27272727
"base_url": "https://generativelanguage.googleapis.com/v1beta/openai/",
2728-
"model": "gemini-2.5-flash",
2728+
"model": "gemini-3.5-flash",
27292729
"key_prefix": None,
27302730
"key_placeholder": "api-key...",
27312731
},
@@ -2735,7 +2735,7 @@ def _handle_prompt_command(
27352735
"key_env": "GROQ_API_KEY",
27362736
"base_env": "GROQ_BASE_URL",
27372737
"base_url": "https://api.groq.com/openai/v1",
2738-
"model": "llama-3.3-70b-versatile",
2738+
"model": "meta-llama/llama-4-maverick-17b-128e-instruct",
27392739
"key_prefix": "gsk_",
27402740
"key_placeholder": "gsk_...",
27412741
},
@@ -2745,7 +2745,7 @@ def _handle_prompt_command(
27452745
"key_env": "DASHSCOPE_API_KEY",
27462746
"base_env": "DASHSCOPE_BASE_URL",
27472747
"base_url": "https://dashscope.aliyuncs.com/compatible-mode/v1",
2748-
"model": "qwen-plus",
2748+
"model": "qwen-plus-latest",
27492749
"key_prefix": "sk-",
27502750
"key_placeholder": "sk-...",
27512751
},
@@ -2755,7 +2755,7 @@ def _handle_prompt_command(
27552755
"key_env": "ZHIPU_API_KEY",
27562756
"base_env": "ZHIPU_BASE_URL",
27572757
"base_url": "https://open.bigmodel.cn/api/paas/v4",
2758-
"model": "glm-4-plus",
2758+
"model": "glm-5.1",
27592759
"key_prefix": None,
27602760
"key_placeholder": "api-key...",
27612761
},
@@ -2765,7 +2765,7 @@ def _handle_prompt_command(
27652765
"key_env": "MOONSHOT_API_KEY",
27662766
"base_env": "MOONSHOT_BASE_URL",
27672767
"base_url": "https://api.moonshot.ai/v1",
2768-
"model": "kimi-k2.5",
2768+
"model": "kimi-k2.6",
27692769
"key_prefix": "sk-",
27702770
"key_placeholder": "sk-...",
27712771
},
@@ -2775,7 +2775,7 @@ def _handle_prompt_command(
27752775
"key_env": "MINIMAX_API_KEY",
27762776
"base_env": "MINIMAX_BASE_URL",
27772777
"base_url": "https://api.minimax.io/v1",
2778-
"model": "MiniMax-Text-01",
2778+
"model": "MiniMax-M2.7",
27792779
"key_prefix": None,
27802780
"key_placeholder": "api-key...",
27812781
},

‎agent/cli/intro.py‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -7,7 +7,7 @@
77
Natural-language finance research that
88
thinks before it answers.
99
10-
Model: deepseek/deepseek-v3.2
10+
Model: deepseek/deepseek-v4-pro
1111
Skills: 72 loaded
1212
Tools: 27 registered
1313
Sessions: 3 prior (use /history to browse)
@@ -68,7 +68,7 @@ def print_banner(
6868
6969
Args:
7070
console: Shared Rich console (use :func:`cli.theme.get_console`).
71-
model: Active LLM model id, e.g. ``"deepseek/deepseek-v3.2"``.
71+
model: Active LLM model id, e.g. ``"deepseek/deepseek-v4-pro"``.
7272
skills: Number of skills currently loaded.
7373
tools: Number of tools registered in the agent registry.
7474
sessions: Number of prior sessions on disk. ``0`` suppresses the

‎agent/cli/onboard.py‎

Lines changed: 8 additions & 8 deletions
Original file line numberDiff line numberDiff line change
@@ -52,20 +52,20 @@ class Provider:
5252

5353
PROVIDERS: Final[tuple[Provider, ...]] = (
5454
Provider("openrouter", "OpenRouter", "recommended — 200+ models, one key",
55-
"deepseek/deepseek-v3.2",
55+
"deepseek/deepseek-v4-pro",
5656
"OPENROUTER_API_KEY", "OPENROUTER_BASE_URL",
5757
"https://openrouter.ai/api/v1", "sk-or-",
58-
("deepseek/deepseek-v3.2", "openai/gpt-5",
59-
"anthropic/claude-opus-4-7", "google/gemini-2.5-pro")),
60-
Provider("openai", "OpenAI", "GPT-5 / GPT-4o direct",
61-
"gpt-4o", "OPENAI_API_KEY", "OPENAI_BASE_URL",
58+
("deepseek/deepseek-v4-pro", "deepseek/deepseek-v4-flash",
59+
"openai/gpt-5.5-pro", "google/gemini-3.5-flash")),
60+
Provider("openai", "OpenAI", "GPT-5.5 direct",
61+
"gpt-5.5-instant", "OPENAI_API_KEY", "OPENAI_BASE_URL",
6262
"https://api.openai.com/v1", "sk-",
63-
("gpt-5", "gpt-4o", "o3-mini")),
63+
("gpt-5.5-instant", "gpt-5.5-pro", "gpt-5.5")),
6464
Provider("deepseek", "DeepSeek",
6565
"cheapest tier — good for batch backtest research",
66-
"deepseek-chat", "DEEPSEEK_API_KEY", "DEEPSEEK_BASE_URL",
66+
"deepseek-v4-pro", "DEEPSEEK_API_KEY", "DEEPSEEK_BASE_URL",
6767
"https://api.deepseek.com/v1", "sk-",
68-
("deepseek-chat", "deepseek-reasoner")),
68+
("deepseek-v4-pro", "deepseek-v4-flash")),
6969
Provider("ollama", "Ollama", "local, free, no API key",
7070
"qwen2.5:32b", None, "OLLAMA_BASE_URL",
7171
"http://localhost:11434", None,

0 commit comments

Comments
 (0)