mirror of
https://github.com/Alishahryar1/free-claude-code.git
synced 2026-07-03 14:05:26 +02:00
Cursor Agent
eabe8db2e8
- Remove _cuda_failed_models and inference-time CPU fallback - auto: try CUDA only, fail fast on RuntimeError (no CPU fallback) - cpu/cuda: use device directly, fail fast on errors - Update docs in config, .env.example, README Co-authored-by: Ali Khokhar <alishahryar2@gmail.com>
63 lines
1.3 KiB
Bash
63 lines
1.3 KiB
Bash
# Provider: "nvidia_nim" | "open_router" | "lmstudio"
|
|
PROVIDER_TYPE=nvidia_nim
|
|
PROVIDER_RATE_LIMIT=40
|
|
PROVIDER_RATE_WINDOW=60
|
|
|
|
|
|
# HTTP client timeouts (seconds) for provider API requests
|
|
HTTP_READ_TIMEOUT=300
|
|
HTTP_WRITE_TIMEOUT=10
|
|
HTTP_CONNECT_TIMEOUT=2
|
|
|
|
|
|
# All Claude model requests are mapped to this model
|
|
MODEL="stepfun-ai/step-3.5-flash"
|
|
|
|
|
|
# NVIDIA NIM Config
|
|
NVIDIA_NIM_API_KEY=""
|
|
|
|
|
|
# OpenRouter Config
|
|
OPENROUTER_API_KEY=""
|
|
|
|
|
|
# LM Studio Config (local provider, no API key required)
|
|
LM_STUDIO_BASE_URL="http://localhost:1234/v1"
|
|
|
|
|
|
# Messaging Platform: "telegram" | "discord"
|
|
MESSAGING_PLATFORM=discord
|
|
MESSAGING_RATE_LIMIT=40
|
|
MESSAGING_RATE_WINDOW=1
|
|
|
|
|
|
# Voice Note Transcription (local faster-whisper; install with: uv sync --extra voice)
|
|
VOICE_NOTE_ENABLED=true
|
|
WHISPER_MODEL=base
|
|
HF_TOKEN=""
|
|
|
|
|
|
# WHISPER_DEVICE: "cpu" | "cuda" | "auto" (auto = try CUDA, fail fast; no fallback)
|
|
WHISPER_DEVICE=cpu
|
|
|
|
|
|
# Telegram Config
|
|
TELEGRAM_BOT_TOKEN=""
|
|
ALLOWED_TELEGRAM_USER_ID=""
|
|
|
|
|
|
# Discord Config
|
|
DISCORD_BOT_TOKEN=""
|
|
ALLOWED_DISCORD_CHANNELS=""
|
|
|
|
|
|
# Agent Config
|
|
CLAUDE_WORKSPACE="./agent_workspace"
|
|
MAX_CLI_SESSIONS=10
|
|
ALLOWED_DIR=""
|
|
FAST_PREFIX_DETECTION=true
|
|
ENABLE_NETWORK_PROBE_MOCK=true
|
|
ENABLE_TITLE_GENERATION_SKIP=true
|
|
ENABLE_SUGGESTION_MODE_SKIP=true
|
|
ENABLE_FILEPATH_EXTRACTION_MOCK=true |