Error listings
fix LiteLLM mid-stream fallback assistant prefill error Claude Includes evidence for LiteLLM troubleshooting demand.
litellm.BadRequestError: AnthropicException - This model does not support assistant message prefill fix Ollama 500 error unable to load model unknown architecture gemma4 Includes evidence for Ollama troubleshooting demand.
Ollama 500 Internal Server Error: unable to load model - unknown model architecture: 'gemma4' fix Ollama SIGSEGV segmentation fault WebP image decode Includes evidence for Ollama troubleshooting demand.
Ollama SIGSEGV: segmentation violation - mtmd_helper_bitmap_init_from_buf decode failure fix Ollama Qwen2.5 14B crash llama runner exit status 2 GPU Includes evidence for Ollama troubleshooting demand.
Ollama 500 Internal Server Error: llama runner process has terminated: signal arrived during cgo execution - Qwen2.5 14B exit status 2 fix Copilot Chat responseBody.getReader is not a function error Includes evidence for GitHub Copilot troubleshooting demand.
responseBody.getReader is not a function: TypeError in VS Code Copilot Chat model provider extension Fix Claude Code VSCode extension 'Unhandled case: [object Object]' crash overlay Includes evidence for Claude Code troubleshooting demand.
Unhandled case: [object Object] — red banner overlay crash in Claude Code VSCode extension Fix LiteLLM proxy Azure OpenAI authentication failure after upgrading to v1.84.0 Includes evidence for LiteLLM troubleshooting demand.
LiteLLM Azure OAI Models Auth Broken in 1.84.0 — enable_azure_ad_token authentication fails after upgrading to v1.84.0 Fix Netlify CLI infinite 'terminated unexpectedly' error loop during ntl dev Includes evidence for Netlify troubleshooting demand.
Netlify CLI 'Error: Netlify CLI has terminated unexpectedly' — infinite error loop after ntl dev Fix Vercel CLI env add storing empty values instead of provided --value Includes evidence for Vercel troubleshooting demand.
vercel env add --value <value> --sensitive --yes silently stores empty string Fix Claude Code prompt cache not lasting 1 hour as expected, causing excessive API costs Includes evidence for Claude Code troubleshooting demand.
Cache TTL silently regressed from 1h to 5m around early March 2026, causing quota and cost inflation Fix Claude Code failing to use Opus 4.7 model through AWS Bedrock provider Includes evidence for Claude Code troubleshooting demand.
Opus 4.7 doesn't work with bedrock — error: model not found or access denied Fix Claude Code unexpectedly using extra usage billing instead of Max plan quota Includes evidence for Claude Code troubleshooting demand.
HERMES.md in git commit messages causes requests to route to extra usage billing instead of plan quota Fix Claude Code creating unnecessarily large cache entries inflating API costs Includes evidence for Claude Code troubleshooting demand.
CC v2.1.100+ inflates cache_creation by ~20K tokens vs v2.1.98 — same payload, server-side Fix Claude Code cache not working when telemetry is disabled Includes evidence for Claude Code troubleshooting demand.
Disabling telemetry also disables 1-hour prompt cache TTL Fix LiteLLM not stripping unsupported parameters before forwarding to Bedrock Includes evidence for LiteLLM troubleshooting demand.
additional_drop_params ignored for POST /v1/messages — Bedrock rejects unsupported fields Fix LiteLLM proxy failing with DeepSeek V4 Pro in multi-turn conversations Includes evidence for LiteLLM troubleshooting demand.
DeepSeek V4 Pro fails in multi-turn conversations — reasoning_content strip Fix LiteLLM self-hosted installation failing at Prisma schema step Includes evidence for LiteLLM troubleshooting demand.
LiteLLM self-hosted installation fail due to schema.prisma error Fix Next.js router state header parse error during version upgrades or rolling deployments Includes evidence for Next.js troubleshooting demand.
The router state header was sent but could not be parsed Fix Turbopack consuming 40+ GB RAM and crashing due to incorrect workspace root detection Includes evidence for Next.js troubleshooting demand.
Next.js Turbopack misdetected workspace root causes RAM exhaustion (40+ GB) and process crash Fix Ollama 500 error on Apple M5 with Metal compiler failure Includes evidence for Ollama troubleshooting demand.
Ollama Apple M5 + macOS 26.3.1: 500 Internal Server Error — Metal compiler failed to build request