llms-py 2.0.7__tar.gz → 2.0.9__tar.gz

{llms_py-2.0.7/llms_py.egg-info → llms_py-2.0.9}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: llms-py
-Version: 2.0.7
+Version: 2.0.9
 Summary: A lightweight CLI tool and OpenAI-compatible server for querying multiple Large Language Model (LLM) providers
 Home-page: https://github.com/ServiceStack/llms
 Author: ServiceStack
@@ -51,7 +51,7 @@ Configure additional providers and models in [llms.json](llms.json)
 ## Features
 
 - **Lightweight**: Single [llms.py](llms.py) Python file with single `aiohttp` dependency
-- **Multi-Provider Support**: OpenRouter, Ollama, Anthropic, Google, OpenAI, Grok, Groq, Qwen, Mistral
+- **Multi-Provider Support**: OpenRouter, Ollama, Anthropic, Google, OpenAI, Grok, Groq, Qwen, Z.ai, Mistral
 - **OpenAI-Compatible API**: Works with any client that supports OpenAI's chat completion API
 - **Configuration Management**: Easy provider enable/disable and configuration management
 - **CLI Interface**: Simple command-line interface for quick interactions
@@ -510,7 +510,52 @@ llms --default grok-4
 
 # Update llms.py to latest version
 llms --update
-```
+
+# Pass custom parameters to chat request (URL-encoded)
+llms --args "temperature=0.7&seed=111" "What is 2+2?"
+
+# Multiple parameters with different types
+llms --args "temperature=0.5&max_completion_tokens=50" "Tell me a joke"
+
+# URL-encoded special characters (stop sequences)
+llms --args "stop=Two,Words" "Count to 5"
+
+# Combine with other options
+llms --system "You are helpful" --args "temperature=0.3" --raw "Hello"
+```
+
+#### Custom Parameters with `--args`
+
+The `--args` option allows you to pass URL-encoded parameters to customize the chat request sent to LLM providers:
+
+**Parameter Types:**
+- **Floats**: `temperature=0.7`, `frequency_penalty=0.2`
+- **Integers**: `max_completion_tokens=100`
+- **Booleans**: `store=true`, `verbose=false`, `logprobs=true`
+- **Strings**: `stop=one`
+- **Lists**: `stop=two,words`
+
+**Common Parameters:**
+- `temperature`: Controls randomness (0.0 to 2.0)
+- `max_completion_tokens`: Maximum tokens in response
+- `seed`: For reproducible outputs
+- `top_p`: Nucleus sampling parameter
+- `stop`: Stop sequences (URL-encode special chars)
+- `store`: Whether or not to store the output
+- `frequency_penalty`: Penalize new tokens based on frequency
+- `presence_penalty`: Penalize new tokens based on presence
+- `logprobs`: Include log probabilities in response
+- `parallel_tool_calls`: Enable parallel tool calls
+- `prompt_cache_key`: Cache key for prompt
+- `reasoning_effort`: Reasoning effort (low, medium, high, *minimal, *none, *default)
+- `safety_identifier`: A string that uniquely identifies each user
+- `seed`: For reproducible outputs
+- `service_tier`: Service tier (free, standard, premium, *default)
+- `top_logprobs`: Number of top logprobs to return
+- `top_p`: Nucleus sampling parameter
+- `verbosity`: Verbosity level (0, 1, 2, 3, *default)
+- `enable_thinking`: Enable thinking mode (Qwen)
+- `stream`: Enable streaming responses
 
 ### Default Model Configuration
 
@@ -558,6 +603,42 @@ llms "Explain quantum computing" | glow
 
 ## Supported Providers
 
+Any OpenAI-compatible providers and their models can be added by configuring them in [llms.json](./llms.json). By default only AI Providers with free tiers are enabled which will only be "available" if their API Key is set. 
+
+You can list the available providers, their models and which are enabled or disabled with:
+
+```bash
+llms ls
+```
+
+They can be enabled/disabled in your `llms.json` file or with:
+
+```bash
+llms --enable <provider>
+llms --disable <provider>
+```
+
+For a provider to be available, they also require their API Key configured in either your Environment Variables
+or directly in your `llms.json`.
+
+### Environment Variables
+
+| Provider        | Variable                  | Description         | Example |
+|-----------------|---------------------------|---------------------|---------|
+| openrouter_free | `OPENROUTER_FREE_API_KEY` | OpenRouter FREE models API key | `sk-or-...` |
+| groq            | `GROQ_API_KEY`            | Groq API key        | `gsk_...` |
+| google_free     | `GOOGLE_FREE_API_KEY`     | Google FREE API key | `AIza...` |
+| codestral       | `CODESTRAL_API_KEY`       | Codestral API key   | `...` |
+| ollama          | N/A                       | No API key required | |
+| openrouter      | `OPENROUTER_API_KEY`      | OpenRouter API key  | `sk-or-...` |
+| google          | `GOOGLE_API_KEY`          | Google API key      | `AIza...` |
+| anthropic       | `ANTHROPIC_API_KEY`       | Anthropic API key   | `sk-ant-...` |
+| openai          | `OPENAI_API_KEY`          | OpenAI API key      | `sk-...` |
+| grok            | `GROK_API_KEY`            | Grok (X.AI) API key | `xai-...` |
+| qwen            | `DASHSCOPE_API_KEY`       | Qwen (Alibaba) API key | `sk-...` |
+| z.ai            | `ZAI_API_KEY`             | Z.ai API key        | `sk-...` |
+| mistral         | `MISTRAL_API_KEY`         | Mistral API key     | `...` |
+
 ### OpenAI
 - **Type**: `OpenAiProvider`
 - **Models**: GPT-5, GPT-5 Codex, GPT-4o, GPT-4o-mini, o3, etc.
@@ -588,6 +669,26 @@ export GOOGLE_API_KEY="your-key"
 llms --enable google_free
 ```
 
+### OpenRouter
+- **Type**: `OpenAiProvider`
+- **Models**: 100+ models from various providers
+- **Features**: Access to latest models, free tier available
+
+```bash
+export OPENROUTER_API_KEY="your-key"
+llms --enable openrouter
+```
+
+### Grok (X.AI)
+- **Type**: `OpenAiProvider`
+- **Models**: Grok-4, Grok-3, Grok-3-mini, Grok-code-fast-1, etc.
+- **Features**: Real-time information, humor, uncensored responses
+
+```bash
+export GROK_API_KEY="your-key"
+llms --enable grok
+```
+
 ### Groq
 - **Type**: `OpenAiProvider`
 - **Models**: Llama 3.3, Gemma 2, Kimi K2, etc.
@@ -608,44 +709,44 @@ llms --enable groq
 llms --enable ollama
 ```
 
-### OpenRouter
+### Qwen (Alibaba Cloud)
 - **Type**: `OpenAiProvider`
-- **Models**: 100+ models from various providers
-- **Features**: Access to latest models, free tier available
+- **Models**: Qwen3-max, Qwen-max, Qwen-plus, Qwen2.5-VL, QwQ-plus, etc.
+- **Features**: Multilingual, vision models, coding, reasoning, audio processing
 
 ```bash
-export OPENROUTER_API_KEY="your-key"
-llms --enable openrouter
+export DASHSCOPE_API_KEY="your-key"
+llms --enable qwen
 ```
 
-### Mistral
+### Z.ai
 - **Type**: `OpenAiProvider`
-- **Models**: Mistral Large, Codestral, Pixtral, etc.
-- **Features**: Code generation, multilingual
+- **Models**: GLM-4.6, GLM-4.5, GLM-4.5-air, GLM-4.5-x, GLM-4.5-airx, GLM-4.5-flash, GLM-4:32b
+- **Features**: Advanced language models with strong reasoning capabilities
 
 ```bash
-export MISTRAL_API_KEY="your-key"
-llms --enable mistral
+export ZAI_API_KEY="your-key"
+llms --enable z.ai
 ```
 
-### Grok (X.AI)
+### Mistral
 - **Type**: `OpenAiProvider`
-- **Models**: Grok-4, Grok-3, Grok-3-mini, Grok-code-fast-1, etc.
-- **Features**: Real-time information, humor, uncensored responses
+- **Models**: Mistral Large, Codestral, Pixtral, etc.
+- **Features**: Code generation, multilingual
 
 ```bash
-export GROK_API_KEY="your-key"
-llms --enable grok
+export MISTRAL_API_KEY="your-key"
+llms --enable mistral
 ```
 
-### Qwen (Alibaba Cloud)
+### Codestral
 - **Type**: `OpenAiProvider`
-- **Models**: Qwen3-max, Qwen-max, Qwen-plus, Qwen2.5-VL, QwQ-plus, etc.
-- **Features**: Multilingual, vision models, coding, reasoning, audio processing
+- **Models**: Codestral
+- **Features**: Code generation
 
 ```bash
-export DASHSCOPE_API_KEY="your-key"
-llms --enable qwen
+export CODESTRAL_API_KEY="your-key"
+llms --enable codestral
 ```
 
 ## Model Routing
@@ -654,22 +755,6 @@ The tool automatically routes requests to the first available provider that supp
 
 Example: If both OpenAI and OpenRouter support `kimi-k2`, the request will first try OpenRouter (free), then fall back to Groq than OpenRouter (Paid) if requests fails.
 
-## Environment Variables
-
-| Variable | Description | Example |
-|----------|-------------|---------|
-| `LLMS_CONFIG_PATH`        | Custom config file path | `/path/to/llms.json` |
-| `OPENAI_API_KEY`          | OpenAI API key | `sk-...` |
-| `ANTHROPIC_API_KEY`       | Anthropic API key | `sk-ant-...` |
-| `GOOGLE_API_KEY`          | Google API key | `AIza...` |
-| `GROQ_API_KEY`            | Groq API key | `gsk_...` |
-| `MISTRAL_API_KEY`         | Mistral API key | `...` |
-| `OPENROUTER_API_KEY`      | OpenRouter API key | `sk-or-...` |
-| `OPENROUTER_FREE_API_KEY` | OpenRouter free tier key | `sk-or-...` |
-| `CODESTRAL_API_KEY`       | Codestral API key | `...` |
-| `GROK_API_KEY`            | Grok (X.AI) API key | `xai-...` |
-| `DASHSCOPE_API_KEY`       | Qwen (Alibaba Cloud) API key | `sk-...` |
-
 ## Configuration Examples
 
 ### Minimal Configuration

{llms_py-2.0.7 → llms_py-2.0.9}/README.md

@@ -11,7 +11,7 @@ Configure additional providers and models in [llms.json](llms.json)
 ## Features
 
 - **Lightweight**: Single [llms.py](llms.py) Python file with single `aiohttp` dependency
-- **Multi-Provider Support**: OpenRouter, Ollama, Anthropic, Google, OpenAI, Grok, Groq, Qwen, Mistral
+- **Multi-Provider Support**: OpenRouter, Ollama, Anthropic, Google, OpenAI, Grok, Groq, Qwen, Z.ai, Mistral
 - **OpenAI-Compatible API**: Works with any client that supports OpenAI's chat completion API
 - **Configuration Management**: Easy provider enable/disable and configuration management
 - **CLI Interface**: Simple command-line interface for quick interactions
@@ -470,7 +470,52 @@ llms --default grok-4
 
 # Update llms.py to latest version
 llms --update
-```
+
+# Pass custom parameters to chat request (URL-encoded)
+llms --args "temperature=0.7&seed=111" "What is 2+2?"
+
+# Multiple parameters with different types
+llms --args "temperature=0.5&max_completion_tokens=50" "Tell me a joke"
+
+# URL-encoded special characters (stop sequences)
+llms --args "stop=Two,Words" "Count to 5"
+
+# Combine with other options
+llms --system "You are helpful" --args "temperature=0.3" --raw "Hello"
+```
+
+#### Custom Parameters with `--args`
+
+The `--args` option allows you to pass URL-encoded parameters to customize the chat request sent to LLM providers:
+
+**Parameter Types:**
+- **Floats**: `temperature=0.7`, `frequency_penalty=0.2`
+- **Integers**: `max_completion_tokens=100`
+- **Booleans**: `store=true`, `verbose=false`, `logprobs=true`
+- **Strings**: `stop=one`
+- **Lists**: `stop=two,words`
+
+**Common Parameters:**
+- `temperature`: Controls randomness (0.0 to 2.0)
+- `max_completion_tokens`: Maximum tokens in response
+- `seed`: For reproducible outputs
+- `top_p`: Nucleus sampling parameter
+- `stop`: Stop sequences (URL-encode special chars)
+- `store`: Whether or not to store the output
+- `frequency_penalty`: Penalize new tokens based on frequency
+- `presence_penalty`: Penalize new tokens based on presence
+- `logprobs`: Include log probabilities in response
+- `parallel_tool_calls`: Enable parallel tool calls
+- `prompt_cache_key`: Cache key for prompt
+- `reasoning_effort`: Reasoning effort (low, medium, high, *minimal, *none, *default)
+- `safety_identifier`: A string that uniquely identifies each user
+- `seed`: For reproducible outputs
+- `service_tier`: Service tier (free, standard, premium, *default)
+- `top_logprobs`: Number of top logprobs to return
+- `top_p`: Nucleus sampling parameter
+- `verbosity`: Verbosity level (0, 1, 2, 3, *default)
+- `enable_thinking`: Enable thinking mode (Qwen)
+- `stream`: Enable streaming responses
 
 ### Default Model Configuration
 
@@ -518,6 +563,42 @@ llms "Explain quantum computing" | glow
 
 ## Supported Providers
 
+Any OpenAI-compatible providers and their models can be added by configuring them in [llms.json](./llms.json). By default only AI Providers with free tiers are enabled which will only be "available" if their API Key is set. 
+
+You can list the available providers, their models and which are enabled or disabled with:
+
+```bash
+llms ls
+```
+
+They can be enabled/disabled in your `llms.json` file or with:
+
+```bash
+llms --enable <provider>
+llms --disable <provider>
+```
+
+For a provider to be available, they also require their API Key configured in either your Environment Variables
+or directly in your `llms.json`.
+
+### Environment Variables
+
+| Provider        | Variable                  | Description         | Example |
+|-----------------|---------------------------|---------------------|---------|
+| openrouter_free | `OPENROUTER_FREE_API_KEY` | OpenRouter FREE models API key | `sk-or-...` |
+| groq            | `GROQ_API_KEY`            | Groq API key        | `gsk_...` |
+| google_free     | `GOOGLE_FREE_API_KEY`     | Google FREE API key | `AIza...` |
+| codestral       | `CODESTRAL_API_KEY`       | Codestral API key   | `...` |
+| ollama          | N/A                       | No API key required | |
+| openrouter      | `OPENROUTER_API_KEY`      | OpenRouter API key  | `sk-or-...` |
+| google          | `GOOGLE_API_KEY`          | Google API key      | `AIza...` |
+| anthropic       | `ANTHROPIC_API_KEY`       | Anthropic API key   | `sk-ant-...` |
+| openai          | `OPENAI_API_KEY`          | OpenAI API key      | `sk-...` |
+| grok            | `GROK_API_KEY`            | Grok (X.AI) API key | `xai-...` |
+| qwen            | `DASHSCOPE_API_KEY`       | Qwen (Alibaba) API key | `sk-...` |
+| z.ai            | `ZAI_API_KEY`             | Z.ai API key        | `sk-...` |
+| mistral         | `MISTRAL_API_KEY`         | Mistral API key     | `...` |
+
 ### OpenAI
 - **Type**: `OpenAiProvider`
 - **Models**: GPT-5, GPT-5 Codex, GPT-4o, GPT-4o-mini, o3, etc.
@@ -548,6 +629,26 @@ export GOOGLE_API_KEY="your-key"
 llms --enable google_free
 ```
 
+### OpenRouter
+- **Type**: `OpenAiProvider`
+- **Models**: 100+ models from various providers
+- **Features**: Access to latest models, free tier available
+
+```bash
+export OPENROUTER_API_KEY="your-key"
+llms --enable openrouter
+```
+
+### Grok (X.AI)
+- **Type**: `OpenAiProvider`
+- **Models**: Grok-4, Grok-3, Grok-3-mini, Grok-code-fast-1, etc.
+- **Features**: Real-time information, humor, uncensored responses
+
+```bash
+export GROK_API_KEY="your-key"
+llms --enable grok
+```
+
 ### Groq
 - **Type**: `OpenAiProvider`
 - **Models**: Llama 3.3, Gemma 2, Kimi K2, etc.
@@ -568,44 +669,44 @@ llms --enable groq
 llms --enable ollama
 ```
 
-### OpenRouter
+### Qwen (Alibaba Cloud)
 - **Type**: `OpenAiProvider`
-- **Models**: 100+ models from various providers
-- **Features**: Access to latest models, free tier available
+- **Models**: Qwen3-max, Qwen-max, Qwen-plus, Qwen2.5-VL, QwQ-plus, etc.
+- **Features**: Multilingual, vision models, coding, reasoning, audio processing
 
 ```bash
-export OPENROUTER_API_KEY="your-key"
-llms --enable openrouter
+export DASHSCOPE_API_KEY="your-key"
+llms --enable qwen
 ```
 
-### Mistral
+### Z.ai
 - **Type**: `OpenAiProvider`
-- **Models**: Mistral Large, Codestral, Pixtral, etc.
-- **Features**: Code generation, multilingual
+- **Models**: GLM-4.6, GLM-4.5, GLM-4.5-air, GLM-4.5-x, GLM-4.5-airx, GLM-4.5-flash, GLM-4:32b
+- **Features**: Advanced language models with strong reasoning capabilities
 
 ```bash
-export MISTRAL_API_KEY="your-key"
-llms --enable mistral
+export ZAI_API_KEY="your-key"
+llms --enable z.ai
 ```
 
-### Grok (X.AI)
+### Mistral
 - **Type**: `OpenAiProvider`
-- **Models**: Grok-4, Grok-3, Grok-3-mini, Grok-code-fast-1, etc.
-- **Features**: Real-time information, humor, uncensored responses
+- **Models**: Mistral Large, Codestral, Pixtral, etc.
+- **Features**: Code generation, multilingual
 
 ```bash
-export GROK_API_KEY="your-key"
-llms --enable grok
+export MISTRAL_API_KEY="your-key"
+llms --enable mistral
 ```
 
-### Qwen (Alibaba Cloud)
+### Codestral
 - **Type**: `OpenAiProvider`
-- **Models**: Qwen3-max, Qwen-max, Qwen-plus, Qwen2.5-VL, QwQ-plus, etc.
-- **Features**: Multilingual, vision models, coding, reasoning, audio processing
+- **Models**: Codestral
+- **Features**: Code generation
 
 ```bash
-export DASHSCOPE_API_KEY="your-key"
-llms --enable qwen
+export CODESTRAL_API_KEY="your-key"
+llms --enable codestral
 ```
 
 ## Model Routing
@@ -614,22 +715,6 @@ The tool automatically routes requests to the first available provider that supp
 
 Example: If both OpenAI and OpenRouter support `kimi-k2`, the request will first try OpenRouter (free), then fall back to Groq than OpenRouter (Paid) if requests fails.
 
-## Environment Variables
-
-| Variable | Description | Example |
-|----------|-------------|---------|
-| `LLMS_CONFIG_PATH`        | Custom config file path | `/path/to/llms.json` |
-| `OPENAI_API_KEY`          | OpenAI API key | `sk-...` |
-| `ANTHROPIC_API_KEY`       | Anthropic API key | `sk-ant-...` |
-| `GOOGLE_API_KEY`          | Google API key | `AIza...` |
-| `GROQ_API_KEY`            | Groq API key | `gsk_...` |
-| `MISTRAL_API_KEY`         | Mistral API key | `...` |
-| `OPENROUTER_API_KEY`      | OpenRouter API key | `sk-or-...` |
-| `OPENROUTER_FREE_API_KEY` | OpenRouter free tier key | `sk-or-...` |
-| `CODESTRAL_API_KEY`       | Codestral API key | `...` |
-| `GROK_API_KEY`            | Grok (X.AI) API key | `xai-...` |
-| `DASHSCOPE_API_KEY`       | Qwen (Alibaba Cloud) API key | `sk-...` |
-
 ## Configuration Examples
 
 ### Minimal Configuration

{llms_py-2.0.7 → llms_py-2.0.9}/llms.json

@@ -9,7 +9,12 @@
             "messages": [
                 {
                     "role": "user",
-                    "content": ""
+                    "content": [
+                        {
+                            "type": "text",
+                            "text": ""
+                        }
+                    ]
                 }
             ]
         },
@@ -193,6 +198,9 @@
                 "nova-micro": "amazon/nova-micro-v1",
                 "nova-lite": "amazon/nova-lite-v1",
                 "nova-pro": "amazon/nova-pro-v1",
+                "claude-opus-4-1": "anthropic/claude-opus-4.1",
+                "claude-sonnet-4-5": "anthropic/claude-sonnet-4.5",
+                "claude-sonnet-4-0": "anthropic/claude-sonnet-4",
                 "gpt-5": "openai/gpt-5",
                 "gpt-5-chat": "openai/gpt-5-chat",
                 "gpt-5-mini": "openai/gpt-5-mini",
@@ -210,11 +218,13 @@
                 "grok-4": "x-ai/grok-4",
                 "grok-4-fast": "x-ai/grok-4-fast",
                 "grok-code-fast-1": "x-ai/grok-code-fast-1",
+                "glm-4.6": "z-ai/glm-4.6",
                 "glm-4.5v": "z-ai/glm-4.5v",
                 "glm-4.5": "z-ai/glm-4.5",
                 "glm-4.5-air": "z-ai/glm-4.5-air",
                 "kimi-k2": "moonshotai/kimi-k2",
                 "deepseek-v3.1:671b": "deepseek/deepseek-chat",
+                "deepseek-v3.2-exp": "deepseek/deepseek-v3.2-exp",
                 "deepseek-chat-v3.1:671b": "deepseek/deepseek-chat-v3.1:free",
                 "deepseek-r1:671b": "deepseek/deepseek-r1",
                 "deepseek-v3.1-terminus": "deepseek/deepseek-v3.1-terminus",
@@ -263,6 +273,7 @@
             "models": {
                 "claude-opus-4-1": "claude-opus-4-1",
                 "claude-opus-4": "claude-opus-4",
+                "claude-sonnet-4-5": "claude-sonnet-4-5",
                 "claude-sonnet-4-0": "claude-sonnet-4-0",
                 "claude-3-7-sonnet": "claude-3-7-sonnet-latest",
                 "claude-3-5-haiku": "claude-3-5-haiku-latest",
@@ -383,7 +394,24 @@
                 "qwen2.5-vl:7b": "qwen2.5-vl-7b-instruct",
                 "qwen2.5-vl:3b": "qwen2.5-vl-3b-instruct",
                 "qwen2.5-omni:7b": "qwen2.5-omni-7b"
-            }
+            },
+            "enable_thinking": false
+        },
+        "z.ai": {
+            "enabled": false,
+            "type": "OpenAiProvider",
+            "base_url": "https://api.z.ai/api/paas/v4",
+            "api_key": "$ZAI_API_KEY",
+            "models": {
+                "glm-4.6": "glm-4.6",
+                "glm-4.5": "glm-4.5",
+                "glm-4.5-air": "glm-4.5-air",
+                "glm-4.5-x": "glm-4.5-x",
+                "glm-4.5-airx": "glm-4.5-airx",
+                "glm-4.5-flash": "glm-4.5-flash",
+                "glm-4:32b": "glm-4-32b-0414-128k"
+            },
+            "temperature": 0.7
         },
         "mistral": {
             "enabled": false,
@@ -396,20 +424,22 @@
                 "devstral-medium": "devstral-medium-2507",
                 "codestral:22b": "codestral-latest",
                 "mistral-ocr": "mistral-ocr-latest",
-                "voxtral-mini": "voxtral-mini-latest",
                 "mistral-small3.2:24b": "mistral-small-latest",
                 "magistral-small": "magistral-small-latest",
                 "devstral-small": "devstral-small-2507",
                 "voxtral-small": "voxtral-small-latest",
+                "voxtral-mini": "voxtral-mini-latest",
+                "codestral-embed": "codestral-embed-2505",
+                "mistral-embed": "mistral-embed",
                 "mistral-large:123b": "mistral-large-latest",
                 "pixtral-large:124b": "pixtral-large-latest",
                 "pixtral:12b": "pixtral-12b",
-                "mistral-nemo:12b": "mistral-nemo",
+                "mistral-nemo:12b": "open-mistral-nemo",
                 "mistral-saba": "mistral-saba-latest",
                 "mistral:7b": "open-mistral-7b",
                 "mixtral:8x7b": "open-mixtral-8x7b",
                 "mixtral:8x22b": "open-mixtral-8x22b",
-                "ministral:8b": "ministral-3b-latest",
+                "ministral:8b": "ministral-8b-latest",
                 "ministral:3b": "ministral-3b-latest"
             }
         }

{llms_py-2.0.7 → llms_py-2.0.9}/llms.py

@@ -14,6 +14,7 @@ import mimetypes
 import traceback
 import sys
 import site
+from urllib.parse import parse_qs
 
 import aiohttp
 from aiohttp import web
@@ -21,7 +22,7 @@ from aiohttp import web
 from pathlib import Path
 from importlib import resources   # Py≥3.9  (pip install importlib_resources for 3.7/3.8)
 
-VERSION = "2.0.7"
+VERSION = "2.0.9"
 _ROOT = None
 g_config_path = None
 g_ui_path = None
@@ -63,7 +64,8 @@ def chat_summary(chat):
                     elif 'file' in item:
                         if 'file_data' in item['file']:
                             data = item['file']['file_data']
-                            item['file']['file_data'] = f"({len(data)})"
+                            prefix = url.split(',', 1)[0]
+                            item['file']['file_data'] = prefix + f",({len(url) - len(prefix)})"
     return json.dumps(clone, indent=2)
 
 def gemini_chat_summary(gemini_chat):
@@ -89,6 +91,60 @@ def is_url(url):
 def get_filename(file):
     return file.rsplit('/',1)[1] if '/' in file else 'file'
 
+def parse_args_params(args_str):
+    """Parse URL-encoded parameters and return a dictionary."""
+    if not args_str:
+        return {}
+
+    # Parse the URL-encoded string
+    parsed = parse_qs(args_str, keep_blank_values=True)
+
+    # Convert to simple dict with single values (not lists)
+    result = {}
+    for key, values in parsed.items():
+        if len(values) == 1:
+            value = values[0]
+            # Try to convert to appropriate types
+            if value.lower() == 'true':
+                result[key] = True
+            elif value.lower() == 'false':
+                result[key] = False
+            elif value.isdigit():
+                result[key] = int(value)
+            else:
+                try:
+                    # Try to parse as float
+                    result[key] = float(value)
+                except ValueError:
+                    # Keep as string
+                    result[key] = value
+        else:
+            # Multiple values, keep as list
+            result[key] = values
+
+    return result
+
+def apply_args_to_chat(chat, args_params):
+    """Apply parsed arguments to the chat request."""
+    if not args_params:
+        return chat
+
+    # Apply each parameter to the chat request
+    for key, value in args_params.items():
+        if isinstance(value, str):
+            if key == 'stop':
+                if ',' in value:
+                    value = value.split(',')
+            elif key == 'max_completion_tokens' or key == 'max_tokens' or key == 'n' or key == 'seed' or key == 'top_logprobs':
+                value = int(value)
+            elif key == 'temperature' or key == 'top_p' or key == 'frequency_penalty' or key == 'presence_penalty':
+                value = float(value)
+            elif key == 'store' or key == 'logprobs' or key == 'enable_thinking' or key == 'parallel_tool_calls' or key == 'stream':
+                value = bool(value)
+        chat[key] = value
+
+    return chat
+
 def is_base_64(data):
     try:
         base64.b64decode(data)
@@ -190,8 +246,9 @@ async def process_chat(chat):
                                     content = f.read()
                                     file['filename'] = get_filename(url)
                                     file['file_data'] = f"data:{mimetype};base64,{base64.b64encode(content).decode('utf-8')}"
-                            elif is_base_64(url):
-                                file['filename'] = 'file'
+                            elif url.startswith('data:'):
+                                if 'filename' not in file:
+                                    file['filename'] = 'file' 
                                 pass # use base64 data as-is
                             else:
                                 raise Exception(f"Invalid file: {url}")
@@ -219,13 +276,38 @@ class OpenAiProvider:
         self.api_key = api_key
         self.models = models
 
-        self.chat_url = f"{base_url}/v1/chat/completions"
+        # check if base_url ends with /v{\d} to handle providers with different versions (e.g. z.ai uses /v4)
+        last_segment = base_url.rsplit('/',1)[1]
+        if last_segment.startswith('v') and last_segment[1:].isdigit():
+            self.chat_url = f"{base_url}/chat/completions"
+        else:
+            self.chat_url = f"{base_url}/v1/chat/completions"
+
         self.headers = kwargs['headers'] if 'headers' in kwargs else {
             "Content-Type": "application/json",
         }
         if api_key is not None:
             self.headers["Authorization"] = f"Bearer {api_key}"
 
+        self.frequency_penalty = float(kwargs['frequency_penalty']) if 'frequency_penalty' in kwargs else None
+        self.max_completion_tokens = int(kwargs['max_completion_tokens']) if 'max_completion_tokens' in kwargs else None
+        self.n = int(kwargs['n']) if 'n' in kwargs else None
+        self.parallel_tool_calls = bool(kwargs['parallel_tool_calls']) if 'parallel_tool_calls' in kwargs else None
+        self.presence_penalty = float(kwargs['presence_penalty']) if 'presence_penalty' in kwargs else None
+        self.prompt_cache_key = kwargs['prompt_cache_key'] if 'prompt_cache_key' in kwargs else None
+        self.reasoning_effort = kwargs['reasoning_effort'] if 'reasoning_effort' in kwargs else None
+        self.safety_identifier = kwargs['safety_identifier'] if 'safety_identifier' in kwargs else None        
+        self.seed = int(kwargs['seed']) if 'seed' in kwargs else None
+        self.service_tier = kwargs['service_tier'] if 'service_tier' in kwargs else None
+        self.stop = kwargs['stop'] if 'stop' in kwargs else None
+        self.store = bool(kwargs['store']) if 'store' in kwargs else None
+        self.temperature = float(kwargs['temperature']) if 'temperature' in kwargs else None
+        self.top_logprobs = int(kwargs['top_logprobs']) if 'top_logprobs' in kwargs else None
+        self.top_p = float(kwargs['top_p']) if 'top_p' in kwargs else None
+        self.verbosity = kwargs['verbosity'] if 'verbosity' in kwargs else None
+        self.stream = bool(kwargs['stream']) if 'stream' in kwargs else None
+        self.enable_thinking = bool(kwargs['enable_thinking']) if 'enable_thinking' in kwargs else None
+
     @classmethod
     def test(cls, base_url=None, api_key=None, models={}, **kwargs):
         return base_url is not None and api_key is not None and len(models) > 0
@@ -241,6 +323,41 @@ class OpenAiProvider:
         # with open(os.path.join(os.path.dirname(__file__), 'chat.wip.json'), "w") as f:
         #     f.write(json.dumps(chat, indent=2))
 
+        if self.frequency_penalty is not None:
+            chat['frequency_penalty'] = self.frequency_penalty
+        if self.max_completion_tokens is not None:
+            chat['max_completion_tokens'] = self.max_completion_tokens
+        if self.n is not None:
+            chat['n'] = self.n
+        if self.parallel_tool_calls is not None:
+            chat['parallel_tool_calls'] = self.parallel_tool_calls
+        if self.presence_penalty is not None:
+            chat['presence_penalty'] = self.presence_penalty
+        if self.prompt_cache_key is not None:
+            chat['prompt_cache_key'] = self.prompt_cache_key
+        if self.reasoning_effort is not None:
+            chat['reasoning_effort'] = self.reasoning_effort
+        if self.safety_identifier is not None:
+            chat['safety_identifier'] = self.safety_identifier
+        if self.seed is not None:
+            chat['seed'] = self.seed
+        if self.service_tier is not None:
+            chat['service_tier'] = self.service_tier
+        if self.stop is not None:
+            chat['stop'] = self.stop
+        if self.store is not None:
+            chat['store'] = self.store
+        if self.temperature is not None:
+            chat['temperature'] = self.temperature
+        if self.top_logprobs is not None:
+            chat['top_logprobs'] = self.top_logprobs
+        if self.top_p is not None:
+            chat['top_p'] = self.top_p
+        if self.verbosity is not None:
+            chat['verbosity'] = self.verbosity
+        if self.enable_thinking is not None:
+            chat['enable_thinking'] = self.enable_thinking
+
         chat = await process_chat(chat)
         _log(f"POST {self.chat_url}")
         _log(chat_summary(chat))
@@ -531,10 +648,14 @@ async def chat_completion(chat):
     # If we get here, all providers failed
     raise first_exception
 
-async def cli_chat(chat, image=None, audio=None, file=None, raw=False):
+async def cli_chat(chat, image=None, audio=None, file=None, args=None, raw=False):
     if g_default_model:
         chat['model'] = g_default_model
 
+    # Apply args parameters to chat request
+    if args:
+        chat = apply_args_to_chat(chat, args)
+
     # process_chat downloads the image, just adding the reference here
     if image is not None:
         first_message = None
@@ -919,6 +1040,7 @@ def main():
     parser.add_argument('--image',        default=None, help='Image input to use in chat completion')
     parser.add_argument('--audio',        default=None, help='Audio input to use in chat completion')
     parser.add_argument('--file',         default=None, help='File input to use in chat completion')
+    parser.add_argument('--args',         default=None, help='URL-encoded parameters to add to chat request (e.g. "temperature=0.7&seed=111")', metavar='PARAMS')
     parser.add_argument('--raw',          action='store_true', help='Return raw AI JSON response')
 
     parser.add_argument('--list',         action='store_true', help='Show list of enabled providers and their models (alias ls provider?)')
@@ -1250,13 +1372,21 @@ def main():
             if len(extra_args) > 0:
                 prompt = ' '.join(extra_args)
                 # replace content of last message if exists, else add
-                last_msg = chat['messages'][-1]
-                if last_msg['role'] == 'user':
-                    last_msg['content'] = prompt
+                last_msg = chat['messages'][-1] if 'messages' in chat else None
+                if last_msg and last_msg['role'] == 'user':
+                    if isinstance(last_msg['content'], list):
+                        last_msg['content'][-1]['text'] = prompt
+                    else:
+                        last_msg['content'] = prompt
                 else:
                     chat['messages'].append({'role': 'user', 'content': prompt})
 
-            asyncio.run(cli_chat(chat, image=cli_args.image, audio=cli_args.audio, file=cli_args.file, raw=cli_args.raw))
+            # Parse args parameters if provided
+            args = None
+            if cli_args.args is not None:
+                args = parse_args_params(cli_args.args)
+
+            asyncio.run(cli_chat(chat, image=cli_args.image, audio=cli_args.audio, file=cli_args.file, args=args, raw=cli_args.raw))
             exit(0)
         except Exception as e:
             print(f"{cli_args.logprefix}Error: {e}")

{llms_py-2.0.7 → llms_py-2.0.9/llms_py.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: llms-py
-Version: 2.0.7
+Version: 2.0.9
 Summary: A lightweight CLI tool and OpenAI-compatible server for querying multiple Large Language Model (LLM) providers
 Home-page: https://github.com/ServiceStack/llms
 Author: ServiceStack
@@ -51,7 +51,7 @@ Configure additional providers and models in [llms.json](llms.json)
 ## Features
 
 - **Lightweight**: Single [llms.py](llms.py) Python file with single `aiohttp` dependency
-- **Multi-Provider Support**: OpenRouter, Ollama, Anthropic, Google, OpenAI, Grok, Groq, Qwen, Mistral
+- **Multi-Provider Support**: OpenRouter, Ollama, Anthropic, Google, OpenAI, Grok, Groq, Qwen, Z.ai, Mistral
 - **OpenAI-Compatible API**: Works with any client that supports OpenAI's chat completion API
 - **Configuration Management**: Easy provider enable/disable and configuration management
 - **CLI Interface**: Simple command-line interface for quick interactions
@@ -510,7 +510,52 @@ llms --default grok-4
 
 # Update llms.py to latest version
 llms --update
-```
+
+# Pass custom parameters to chat request (URL-encoded)
+llms --args "temperature=0.7&seed=111" "What is 2+2?"
+
+# Multiple parameters with different types
+llms --args "temperature=0.5&max_completion_tokens=50" "Tell me a joke"
+
+# URL-encoded special characters (stop sequences)
+llms --args "stop=Two,Words" "Count to 5"
+
+# Combine with other options
+llms --system "You are helpful" --args "temperature=0.3" --raw "Hello"
+```
+
+#### Custom Parameters with `--args`
+
+The `--args` option allows you to pass URL-encoded parameters to customize the chat request sent to LLM providers:
+
+**Parameter Types:**
+- **Floats**: `temperature=0.7`, `frequency_penalty=0.2`
+- **Integers**: `max_completion_tokens=100`
+- **Booleans**: `store=true`, `verbose=false`, `logprobs=true`
+- **Strings**: `stop=one`
+- **Lists**: `stop=two,words`
+
+**Common Parameters:**
+- `temperature`: Controls randomness (0.0 to 2.0)
+- `max_completion_tokens`: Maximum tokens in response
+- `seed`: For reproducible outputs
+- `top_p`: Nucleus sampling parameter
+- `stop`: Stop sequences (URL-encode special chars)
+- `store`: Whether or not to store the output
+- `frequency_penalty`: Penalize new tokens based on frequency
+- `presence_penalty`: Penalize new tokens based on presence
+- `logprobs`: Include log probabilities in response
+- `parallel_tool_calls`: Enable parallel tool calls
+- `prompt_cache_key`: Cache key for prompt
+- `reasoning_effort`: Reasoning effort (low, medium, high, *minimal, *none, *default)
+- `safety_identifier`: A string that uniquely identifies each user
+- `seed`: For reproducible outputs
+- `service_tier`: Service tier (free, standard, premium, *default)
+- `top_logprobs`: Number of top logprobs to return
+- `top_p`: Nucleus sampling parameter
+- `verbosity`: Verbosity level (0, 1, 2, 3, *default)
+- `enable_thinking`: Enable thinking mode (Qwen)
+- `stream`: Enable streaming responses
 
 ### Default Model Configuration
 
@@ -558,6 +603,42 @@ llms "Explain quantum computing" | glow
 
 ## Supported Providers
 
+Any OpenAI-compatible providers and their models can be added by configuring them in [llms.json](./llms.json). By default only AI Providers with free tiers are enabled which will only be "available" if their API Key is set. 
+
+You can list the available providers, their models and which are enabled or disabled with:
+
+```bash
+llms ls
+```
+
+They can be enabled/disabled in your `llms.json` file or with:
+
+```bash
+llms --enable <provider>
+llms --disable <provider>
+```
+
+For a provider to be available, they also require their API Key configured in either your Environment Variables
+or directly in your `llms.json`.
+
+### Environment Variables
+
+| Provider        | Variable                  | Description         | Example |
+|-----------------|---------------------------|---------------------|---------|
+| openrouter_free | `OPENROUTER_FREE_API_KEY` | OpenRouter FREE models API key | `sk-or-...` |
+| groq            | `GROQ_API_KEY`            | Groq API key        | `gsk_...` |
+| google_free     | `GOOGLE_FREE_API_KEY`     | Google FREE API key | `AIza...` |
+| codestral       | `CODESTRAL_API_KEY`       | Codestral API key   | `...` |
+| ollama          | N/A                       | No API key required | |
+| openrouter      | `OPENROUTER_API_KEY`      | OpenRouter API key  | `sk-or-...` |
+| google          | `GOOGLE_API_KEY`          | Google API key      | `AIza...` |
+| anthropic       | `ANTHROPIC_API_KEY`       | Anthropic API key   | `sk-ant-...` |
+| openai          | `OPENAI_API_KEY`          | OpenAI API key      | `sk-...` |
+| grok            | `GROK_API_KEY`            | Grok (X.AI) API key | `xai-...` |
+| qwen            | `DASHSCOPE_API_KEY`       | Qwen (Alibaba) API key | `sk-...` |
+| z.ai            | `ZAI_API_KEY`             | Z.ai API key        | `sk-...` |
+| mistral         | `MISTRAL_API_KEY`         | Mistral API key     | `...` |
+
 ### OpenAI
 - **Type**: `OpenAiProvider`
 - **Models**: GPT-5, GPT-5 Codex, GPT-4o, GPT-4o-mini, o3, etc.
@@ -588,6 +669,26 @@ export GOOGLE_API_KEY="your-key"
 llms --enable google_free
 ```
 
+### OpenRouter
+- **Type**: `OpenAiProvider`
+- **Models**: 100+ models from various providers
+- **Features**: Access to latest models, free tier available
+
+```bash
+export OPENROUTER_API_KEY="your-key"
+llms --enable openrouter
+```
+
+### Grok (X.AI)
+- **Type**: `OpenAiProvider`
+- **Models**: Grok-4, Grok-3, Grok-3-mini, Grok-code-fast-1, etc.
+- **Features**: Real-time information, humor, uncensored responses
+
+```bash
+export GROK_API_KEY="your-key"
+llms --enable grok
+```
+
 ### Groq
 - **Type**: `OpenAiProvider`
 - **Models**: Llama 3.3, Gemma 2, Kimi K2, etc.
@@ -608,44 +709,44 @@ llms --enable groq
 llms --enable ollama
 ```
 
-### OpenRouter
+### Qwen (Alibaba Cloud)
 - **Type**: `OpenAiProvider`
-- **Models**: 100+ models from various providers
-- **Features**: Access to latest models, free tier available
+- **Models**: Qwen3-max, Qwen-max, Qwen-plus, Qwen2.5-VL, QwQ-plus, etc.
+- **Features**: Multilingual, vision models, coding, reasoning, audio processing
 
 ```bash
-export OPENROUTER_API_KEY="your-key"
-llms --enable openrouter
+export DASHSCOPE_API_KEY="your-key"
+llms --enable qwen
 ```
 
-### Mistral
+### Z.ai
 - **Type**: `OpenAiProvider`
-- **Models**: Mistral Large, Codestral, Pixtral, etc.
-- **Features**: Code generation, multilingual
+- **Models**: GLM-4.6, GLM-4.5, GLM-4.5-air, GLM-4.5-x, GLM-4.5-airx, GLM-4.5-flash, GLM-4:32b
+- **Features**: Advanced language models with strong reasoning capabilities
 
 ```bash
-export MISTRAL_API_KEY="your-key"
-llms --enable mistral
+export ZAI_API_KEY="your-key"
+llms --enable z.ai
 ```
 
-### Grok (X.AI)
+### Mistral
 - **Type**: `OpenAiProvider`
-- **Models**: Grok-4, Grok-3, Grok-3-mini, Grok-code-fast-1, etc.
-- **Features**: Real-time information, humor, uncensored responses
+- **Models**: Mistral Large, Codestral, Pixtral, etc.
+- **Features**: Code generation, multilingual
 
 ```bash
-export GROK_API_KEY="your-key"
-llms --enable grok
+export MISTRAL_API_KEY="your-key"
+llms --enable mistral
 ```
 
-### Qwen (Alibaba Cloud)
+### Codestral
 - **Type**: `OpenAiProvider`
-- **Models**: Qwen3-max, Qwen-max, Qwen-plus, Qwen2.5-VL, QwQ-plus, etc.
-- **Features**: Multilingual, vision models, coding, reasoning, audio processing
+- **Models**: Codestral
+- **Features**: Code generation
 
 ```bash
-export DASHSCOPE_API_KEY="your-key"
-llms --enable qwen
+export CODESTRAL_API_KEY="your-key"
+llms --enable codestral
 ```
 
 ## Model Routing
@@ -654,22 +755,6 @@ The tool automatically routes requests to the first available provider that supp
 
 Example: If both OpenAI and OpenRouter support `kimi-k2`, the request will first try OpenRouter (free), then fall back to Groq than OpenRouter (Paid) if requests fails.
 
-## Environment Variables
-
-| Variable | Description | Example |
-|----------|-------------|---------|
-| `LLMS_CONFIG_PATH`        | Custom config file path | `/path/to/llms.json` |
-| `OPENAI_API_KEY`          | OpenAI API key | `sk-...` |
-| `ANTHROPIC_API_KEY`       | Anthropic API key | `sk-ant-...` |
-| `GOOGLE_API_KEY`          | Google API key | `AIza...` |
-| `GROQ_API_KEY`            | Groq API key | `gsk_...` |
-| `MISTRAL_API_KEY`         | Mistral API key | `...` |
-| `OPENROUTER_API_KEY`      | OpenRouter API key | `sk-or-...` |
-| `OPENROUTER_FREE_API_KEY` | OpenRouter free tier key | `sk-or-...` |
-| `CODESTRAL_API_KEY`       | Codestral API key | `...` |
-| `GROK_API_KEY`            | Grok (X.AI) API key | `xai-...` |
-| `DASHSCOPE_API_KEY`       | Qwen (Alibaba Cloud) API key | `sk-...` |
-
 ## Configuration Examples
 
 ### Minimal Configuration

{llms_py-2.0.7 → llms_py-2.0.9}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "llms-py"
-version = "2.0.7"
+version = "2.0.9"
 description = "A lightweight CLI tool and OpenAI-compatible server for querying multiple Large Language Model (LLM) providers"
 readme = "README.md"
 license = "BSD-3-Clause"

{llms_py-2.0.7 → llms_py-2.0.9}/setup.py

@@ -16,7 +16,7 @@ with open(os.path.join(this_directory, "requirements.txt"), encoding="utf-8") as
 
 setup(
     name="llms-py",
-    version="2.0.7",
+    version="2.0.9",
     author="ServiceStack",
     author_email="team@servicestack.net",
     description="A lightweight CLI tool and OpenAI-compatible server for querying multiple Large Language Model (LLM) providers",

{llms_py-2.0.7 → llms_py-2.0.9}/ui/Main.mjs

@@ -20,7 +20,7 @@ const ProviderStatus = {
                     <span class="text-red-700">{{(config.status.disabled||[]).length}}</span>
                 </div>
             </button>
-            <div v-if="showPopover" ref="popoverRef" class="absolute right-0 mt-2 w-72 max-h-112 overflow-y-auto bg-white border border-gray-200 rounded-md shadow-lg z-10">
+            <div v-if="showPopover" ref="popoverRef" class="absolute right-0 mt-2 w-72 max-h-116 overflow-y-auto bg-white border border-gray-200 rounded-md shadow-lg z-10">
                 <div class="divide-y divide-gray-100">
                     <div v-for="p in allProviders" :key="p" class="flex items-center justify-between px-3 py-2">
                         <label :for="'chk_' + p" class="cursor-pointer text-sm text-gray-900 truncate mr-2" :title="p">{{ p }}</label>

{llms_py-2.0.7 → llms_py-2.0.9}/ui/Recents.mjs

@@ -1,5 +1,5 @@
-import { ref, computed, onMounted, watch, inject } from 'vue'
-import { useRouter } from 'vue-router'
+import { ref, onMounted, watch, inject } from 'vue'
+import { useRouter, useRoute } from 'vue-router'
 import { useThreadStore } from './threadStore.mjs'
 import { renderMarkdown } from './markdown.mjs'
 
@@ -169,7 +169,36 @@ export default {
         </div>
     `,
     setup() {
+        const router = useRouter()
+        const route = useRoute()
         const q = ref('')
+
+        // Initialize search query from URL parameter
+        onMounted(() => {
+            const urlQuery = route.query.q || ''
+            q.value = urlQuery
+        })
+
+        // Watch for changes in the search input and update URL
+        watch(q, (newQuery) => {
+            const currentQuery = route.query.q || ''
+            if (newQuery !== currentQuery) {
+                // Update URL without triggering navigation
+                router.replace({
+                    path: route.path,
+                    query: newQuery ? { q: newQuery } : {}
+                })
+            }
+        })
+
+        // Watch for URL changes (browser back/forward) and update search input
+        watch(() => route.query.q, (newQuery) => {
+            const urlQuery = newQuery || ''
+            if (q.value !== urlQuery) {
+                q.value = urlQuery
+            }
+        })
+
         return {
             q,
         }