* test(relayconvert): add golden snapshot matrix and relaykit boundary guard
Phase 0 of the relaykit extraction plan: pin byte-level output of every
registered (from,to) request/response/stream conversion route, and
forbid kit-bound packages from growing host-only imports.
* wip(relayconvert): drop gin.Context from converter signatures; add convmeta draft
Phase 1 in progress: relayconvert now takes context.Context; host media
resolver adapts gin.Context back at the service boundary.
* refactor(relayconvert): decouple converters from RelayInfo, gin, and settings
Phase 1 of the relaykit extraction plan:
- converters now depend on convmeta.Meta (implemented by RelayInfo) instead
of *relaycommon.RelayInfo; ClaudeConvertInfo and the format guesser move
to convmeta with aliases left behind
- host settings reach converters via a convmeta.Options snapshot built in
RelayInfo.ConvOptions; no more model_setting/reasoning global reads inside
the conversion layer
- effort-suffix helpers move to service/relayconvert/reasoning (old package
forwards); chat-to-responses upgrade policy moves to service (host routing
logic, not conversion)
- golden conversion matrix unchanged
* test(relayconvert): tighten boundary — kit packages now free of gin/setting imports
* refactor(dto): drop gin and logger dependencies
Phase 2 (part 1): dto.Request.IsStream now takes *http.Request instead of
*gin.Context (Gemini's impl reads query/path off the std request); dto's
three logger calls become common.SysError. Boundary test allowlist is now
empty — kit-bound packages import no gin/setting/logger/model.
* refactor(kit): extract dependency-free kitutil; dto/types/relayconvert stop importing common
Phase 2 of the relaykit extraction plan:
- new service/relayconvert/kitutil holds the pure helpers the kit needs
(JSON wrappers, pointer/string/uuid/timestamp utils, MaskSensitiveInfo,
pluggable LogInfo/LogError hooks, Debug flag)
- dto, types, and all relayconvert packages now use kitutil; their only
remaining internal deps are dto/types/constant
- common keeps every original symbol (MaskSensitiveInfo delegates to
kitutil) so host code is untouched; main.go routes kit logging into
common.SysLog/SysError and mirrors DebugEnabled
- golden conversion matrix unchanged
* refactor(kit): move EndpointType/FinishReason to types; OpenRouter dialect via Options
Kit packages (dto/types/relayconvert/reasonmap) no longer import constant:
- EndpointType and finish-reason values live in types; constant re-exports
- the OpenRouter special-case in claude->openai request conversion reads
Options.OpenRouterDialect, set by the host from the channel type;
InitChannelMeta invalidates the cached snapshot on channel switch
* refactor: extract relaykit submodule (dto/types/relayconvert/reasonmap)
Phase 3 of the relaykit extraction plan:
- new go module github.com/QuantumNous/new-api/relaykit containing dto
(minus task family), types, relayconvert (with convmeta/kitutil/reasoning),
and reasonmap; host consumes it via require + replace, go.work for dev
- task-family dto (task/suno/midjourney/video) stays in the host dto
package; dual-consumer host files alias it as taskdto
- relaykit builds and tests standalone (GOWORK=off): no host imports,
no gin, no DB, no settings
- golden conversion matrix unchanged
* build(docker): copy relaykit/go.mod before go mod download
The local-replace submodule's go.mod must exist inside the build context
for the main module graph to resolve.
* fix: address relaykit extraction regressions
* fix: address relaykit review regressions
* docs: document Meta nil receiver contract
* fix(relaykit): fail OpenAI→Claude conversion without max_tokens; reject negative default_max_tokens
The Claude Messages API requires max_tokens (omitting it is a 400
"Field required"), but with a nil Options.Claude.DefaultMaxTokens hook
the converters silently emitted a request the upstream is guaranteed to
reject. Both OpenAI Chat and Responses → Claude conversions now return
sharedclaude.ErrMissingMaxTokens when no path (client value, default
hook, thinking-adapter floor) supplied one. Unreachable in the host,
which always configures the hook.
Host side, claude.default_max_tokens now rejects negative values at the
option API before persisting — they would wrap into huge unsigned values
during conversion. Zero stays allowed: the current API treats
max_tokens: 0 as cache pre-warming.
* fix: make Gemini safety settings read path race-free
* refactor: consolidate relay protocol converters
* refactor relayconvert text converters
* feat: refine relay converters and advanced custom routing
* refactor: enhance logging and add thought signature handling for Gemini requests
* refactor: enhance channel cache and pricing endpoint handling for advanced custom models
* feat: preserve billing usage semantics
* feat: add protocol-aware billing usage
* Delete useless files
* chore: update action versions in workflow files
* chore: update Docker action versions in workflow files
* fix: harden billing usage settlement and hot-path route matching
- estimate Gemini completion tokens locally when billable usageMetadata is
prompt-only but output content was received (e.g. client aborts the stream
before the final chunk), and rebuild the attached billing_usage as estimated
so settlement does not bill zero output tokens
- guard NewClaudeMessagesBillingUsage against all-zero ClaudeUsage, matching
the OpenAI/Gemini constructors, so a zero billing_usage cannot override a
non-zero top-level usage during settlement
- cache compiled advanced-custom route model regexes; they run on the request
hot path and were recompiled per request
- move the effectiveBillingUsage remap to PostTextConsumeQuota only, and
document that calculateTextQuotaSummary expects remapped usage
- document the updatePricingLock -> channelSyncLock lock ordering that
InitChannelCache/CacheUpdateChannel rely on, and the aux-struct pitfall in
GeminiChatResponse.UnmarshalJSON
* fix(openai): harden Chat-to-Responses compatibility
Add a shared Responses-to-Chat stream state machine and use it from the OpenAI relay path. Preserve assistant text alongside tool calls, bind tool argument deltas by output_index, map incomplete finish reasons, support reasoning/custom tool events, and buffer upstream SSE for non-stream Chat clients.
Add deterministic service tests and relay SSE tests for the conversion path.
Related to #5745.
* refactor: rename openaicompat to relayconvert for improved clarity
* feat(gemini): support responses request conversion
* feat: add responses to chat conversion support
* fix: harden responses chat conversion edge cases
Three layered optimizations targeting Gemini-style 5MB base64 payloads where
RSS could balloon to tens of GB under concurrent load:
1. Byte-based param override (relay/common/override.go)
- Switch legacy/operations hot paths from common.Marshal round-trips and
map[string]any conversions to gjson/sjson on []byte directly.
- Avoids cloning 5MB strings during each Set/Delete operation.
2. strings.Builder for Gemini response markdown (relay/channel/gemini/relay-gemini.go)
- Replace string concatenation + strings.Join when assembling
"" content for inline image responses.
- Pre-allocates capacity from inline_data byte sizes.
3. Outbound BodyStorage + streaming Decoder (this commit's core)
- New relay/common/outbound_body.go helper wraps marshaled upstream bodies
in common.BodyStorage, allowing disk-cache mode to offload jsonData to
a temp file while waiting for upstream TTFB. The original []byte can
then be GC'd, removing ~5MB/req of heap residency during the longest
window of a request.
- All 7 relay handlers (gemini/claude/responses/embedding/image/compatible/
rerank) plus chat_completions_via_responses adopt the helper with
defer closer.Close() and explicit jsonData = nil.
- relay/common/relay_info.go: new UpstreamRequestBodySize so
relay/channel/api_request.go can populate req.ContentLength (lost when
body becomes a type-erased io.Reader).
- common/gin.go UnmarshalBodyReusable: when storage is disk-backed and
content-type is JSON, decode via DecodeJson(storage) instead of
storage.Bytes()+Unmarshal, removing one transient 5MB copy per request.
memory mode and form/multipart paths unchanged.
Add detection, MIME type mapping, and dimension parsing for HEIC/HEIF
images via ISOBMFF ftyp brand inspection and ispe box parsing. Update
Gemini relay to accept these formats and refactor getImageConfig to
properly retry decoders using buffered data.
- Add StreamStatus type (relay/common) to track stream end reason
(done/timeout/client_gone/scanner_error/eof/panic/ping_fail) and
accumulate soft errors during streaming via sync.Once + sync.Mutex.
- Add StreamResult (relay/helper) as the callback interface: adapters
call sr.Error() for soft errors, sr.Stop() for fatal, sr.Done() for
normal completion. No early-return problem — multiple errors per chunk
are naturally supported.
- Refactor StreamScannerHandler callback from func(string) bool to
func(string, *StreamResult). All 9 channel adapters updated.
- Write stream_status into log other JSON field (admin-only) with
status ok/error, end_reason, error_count, and error messages.
- Frontend: display stream status in log detail expansion for admins.
- Introduced a new rule for the Billing Expression System, emphasizing the importance of reading `pkg/billingexpr/expr.md` for dynamic billing.
- Updated the billing expression logic to support new variables and improved handling of image and audio tokens.
- Enhanced the tiered billing functionality with versioning support for expressions and refined quota calculations.
- Added tests to validate the new billing expression features and ensure correctness in pricing calculations.
Use the native Gemini Models API (/v1beta/models) instead of the OpenAI-compatible
path when listing models for Gemini channels, improving compatibility with
third-party Gemini-format providers that don't implement OpenAI routes.
- Add paginated model listing with timeout and optional proxy support
- Select an enabled key for multi-key Gemini channels
- Introduced new OpenAI text models in `common/model.go`.
- Added `IsOpenAITextModel` function to check for OpenAI text models.
- Refactored token estimation methods across various channels to use estimated prompt tokens instead of direct prompt token counts.
- Updated related functions and structures to accommodate the new token estimation approach, enhancing overall token management.
Calcium-Ion