Fixes#7772
The issue was caused by duplicate registration of the MCP endpoint
/mcp/v1/chat/completions in both openai.go and localai.go, leading
to a race condition where requests would randomly hit different
handlers with incompatible behaviors.
Changes:
- Removed duplicate MCP route registration from openai.go
- Kept the localai.MCPStreamEndpoint as the canonical handler
- Added all three MCP route patterns for backward compatibility:
* /v1/mcp/chat/completions
* /mcp/v1/chat/completions
* /mcp/chat/completions
- Added comments to clarify route ownership and prevent future conflicts
- Fixed formatting in ui_api.go
The localai.MCPStreamEndpoint handler is more feature-complete as it
supports both streaming and non-streaming modes, while the removed
openai.MCPCompletionEndpoint only supported synchronous requests.
This eliminates the ~50% failure rate where the cogito library would
receive "Invalid http method" errors when internal HTTP requests were
routed to the wrong handler.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Signed-off-by: majiayu000 <1835304752@qq.com>
* chore: drop mode from image generation(unused)
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* feat(UI): improve image generation front-end
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* feat(UI): only ref images. files is to be deprecated
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* do not override default steps
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
---------
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* fix: add retry logic and fallback for checksums.txt download
- Add HTTP client with 30s timeout to ReleaseManager
- Implement downloadFileWithRetry with 3 attempts and exponential backoff
- Allow manual checksum placement at ~/.localai/checksums/checksums-<version>.txt
- Continue installation with warning if checksum download/verification fails
- Add test for HTTPClient initialization
- Fix linter error in systray_manager.go
Fixes#7385
Signed-off-by: majiayu000 <1835304752@qq.com>
* fix: add retry logic and improve checksums.txt download handling
This commit addresses issue #7385 by implementing:
- Retry logic (3 attempts) for checksum file downloads
- Fallback to manually placed checksum files
- Option to proceed with installation if checksums unavailable (with warnings)
- Fixed resource leaks in download retry loop
- Added configurable HTTP client with 30s timeout
The installation will now be more resilient to network issues while
maintaining security through checksum verification when available.
Signed-off-by: majiayu000 <1835304752@qq.com>
* fix: check for existing checksum file before downloading
This commit addresses the review feedback from mudler on PR #7788.
The code now checks if there's already a checksum file (either manually
placed or previously downloaded) and honors that, skipping download
entirely in such case.
Changes:
- Check for existing checksum file at ~/.localai/checksums/checksums-<version>.txt first
- Check for existing downloaded checksum file at binary path
- Only attempt to download if no existing checksum file is found
- This prevents unnecessary network requests and honors user-placed checksums
Signed-off-by: majiayu000 <1835304752@qq.com>
🤖 Generated with Claude Code
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
---------
Signed-off-by: majiayu000 <1835304752@qq.com>
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
fix: validate MCP configuration in model config
Fixes#7334
The Validate() function was not checking if MCP configuration
(mcp.stdio and mcp.remote) contains valid JSON. This caused
malformed JSON with missing commas to be silently accepted.
Changes:
- Add MCP configuration validation to ModelConfig.Validate()
- Properly report validation errors instead of discarding them
- Add test cases for valid and invalid MCP configurations
The fix ensures that malformed JSON in MCP config sections
will now be caught and reported during validation.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Signed-off-by: majiayu000 <1835304752@qq.com>
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
* fix: Add usage fields to image generation response for OpenAI API compatibility
Fixes#7354
Added input_tokens, output_tokens, and input_tokens_details fields to the
image generation API response to comply with OpenAI's image generation API
specification. This resolves validation errors in LiteLLM and the OpenAI SDK.
Changes:
- Added InputTokensDetails struct with text_tokens and image_tokens fields
- Extended OpenAIUsage struct with input_tokens, output_tokens, and input_tokens_details
- Updated ImageEndpoint to populate usage object with required fields
- Updated InpaintingEndpoint to populate usage object with required fields
- All fields initialized to 0 as per current behavior
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Signed-off-by: majiayu000 <1835304752@qq.com>
* fix: Correct usage field types for image generation API compatibility
Changed InputTokens and OutputTokens from pointer types (*int) to
regular int types to match OpenAI API specification. This fixes
validation errors with LiteLLM and OpenAI SDK when parsing image
generation responses.
Fixes#7354🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Signed-off-by: majiayu000 <1835304752@qq.com>
---------
Signed-off-by: majiayu000 <1835304752@qq.com>
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
Fixes#7420
Added nil checks before calling mergo.Merge in InstallModelFromGallery and InstallModel
functions to prevent panic when req.Overrides or configOverrides are nil. The panic was
occurring at models.go:248 during Qwen-Image-Edit gallery model download.
Changes:
- Added nil check for req.Overrides before merging in InstallModelFromGallery (line 126)
- Added nil check for configOverrides before merging in InstallModel (line 248)
- Added test case to verify nil configOverrides are handled without panic
Signed-off-by: majiayu000 <1835304752@qq.com>
An example output of `rocm-smi --showproductname --showmeminfo vram --showuniqueid --csv`:
```
device,Unique ID,VRAM Total Memory (B),VRAM Total Used Memory (B),Card Series,Card Model,Card Vendor,Card SKU,Subsystem ID,Device Rev,Node ID,GUID,GFX Version
card0,0x9246____________,17163091968,692142080,Navi 21 [Radeon RX 6800/6800 XT / 6900 XT],0x73bf,Advanced Micro Devices Inc. [AMD/ATI],001,0x2406,0xc1,1,45534,gfx1030
card1,N/A,67108864,26079232,Raphael,0x164e,Advanced Micro Devices Inc. [AMD/ATI],RAPHAEL,0x364e,0xc6,2,52156,gfx1036
```
Total memory is actually showed before the total used memory as can be seen in https://github.com/LostRuins/koboldcpp/issues/1104#issuecomment-2321143507.
This PR fixes https://github.com/mudler/LocalAI/issues/7724
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* feat: allow to set forcing backends eviction while requests are in flight
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* feat: try to make the request sit and retry if eviction couldn't be done
Otherwise calls that in order to pass would need to shutdown other
backends would just fail.
In this way instead we make the request sit and retry eviction until it
succeeds. The thresholds can be configured by the user.
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* add tests
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* expose settings to CLI
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* Update docs
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
---------
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
