0193018af6
Merge pull request #4042 from feitianbubu/pr/fe9713dcbf8795e127fbea2fcb1f3011da86ad54
...
新增seedance2.0视频接口支持
2026-04-02 21:30:31 +08:00
41cd051ea9
Merge pull request #3505 from seefs001/fix/claude-media-support
...
fix: add basic inline file support for Claude relay
2026-04-02 13:29:21 +08:00
c04f82bfb5
TODO: fix chat -> messages file type
2026-04-02 13:16:58 +08:00
dafc7618c3
feat: add seedance fail reason
2026-04-02 12:26:44 +08:00
22692b3f87
feat: seedance support seconds
2026-04-02 12:26:37 +08:00
d36e892905
fix: seedance only one text
2026-04-02 12:26:33 +08:00
3cd1ba4673
fix: seedance metadata override prompt
2026-04-02 12:26:27 +08:00
b7c0f754ad
feat: add seedance2.0 video api
2026-04-02 12:24:24 +08:00
7efb1922fe
Merge pull request #3526 from feitianbubu/pr/e560265b6e57aa7b95bc98cb53397ef0a3082d9d
...
支持wan2.7生图-wan2.7-image
2026-04-02 00:15:04 +08:00
89fe99f3bd
Merge pull request #3512 from imlhb/patch-2
...
fix: prevent double-counting of image count n in billing
2026-04-02 00:14:39 +08:00
e5b5331d3b
feat: wan 2.7 support N for gen images number
2026-04-01 17:39:50 +08:00
18373c6eac
feat: add wan 2.7
2026-04-01 17:39:11 +08:00
ab99c30884
fix: move image count n to OtherRatio to prevent double-counting
...
The previous commit commented out AddOtherRatio("n") in the Ali
adaptor to fix double-counting but this could cause billing evasion
when n is specified via extra["parameters"] instead of request.N.
Root cause: ImagePriceRatio in GetTokenCountMeta() already included
n, AND channel adaptors added OtherRatio("n"), resulting in n²
billing.
Proper fix:
- Remove n from ImagePriceRatio (keep sizeRatio * qualityRatio only)
- In ImageHelper, add default OtherRatio("n") when adaptor hasn't
set one; set fallback tokens to 1 (base unit)
- Restore Ali adaptor's AddOtherRatio("n") — it uses actual upstream
parameters/response count, preventing billing evasion
2026-03-31 23:58:10 +08:00
d22f889e5d
fix(xAI): set MaxTokens to nil when MaxCompletionTokens is 0 for grok-3-mini model
2026-03-31 19:16:16 +08:00
20399d3c8f
fix: harden SSRF protection for unauthenticated and user-level endpoints
...
- Add ValidateURLWithFetchSetting check before fetching MJ image URLs
in RelayMidjourneyImage (unauthenticated endpoint)
- Add ValidateURLWithFetchSetting check before fetching video URLs
in VideoProxy (upstream-controlled URL)
- Enable ApplyIPFilterForDomain by default to prevent DNS rebinding
bypass of SSRF protection
- Elevate FetchModels endpoint from AdminAuth to RootAuth
- Update frontend: mark domain IP filtering as recommended, update
description and i18n translations (zh-CN/zh-TW/en/fr/ja/ru/vi)
2026-03-31 17:57:47 +08:00
53aeee4ff7
Comment out price data adjustment logic
...
Comment out code that modifies price data based on image count.
测试发现，如果是接入阿里百炼平台的qwen-image-2.0系列模型，这边计费的时候会出现 0.2*n*倍率*n的情况，最前面的0.2*n会直接显示为模型价格。
例如：
日志详情 模型价格 $0.600000，专属倍率 0.3
其他详情 大小 1080*1920, 品质 standard, 生成数量 3, 其他倍率 n: 3.000000
计费过程 模型价格：$0.600000 * 专属倍率：0.3 = $0.180000
2026-03-31 17:12:06 +08:00
5238f279db
feat: record stream interruption reasons via StreamStatus
...
- 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.
2026-03-31 16:54:39 +08:00
263b9bc695
fix: add basic inline file support for Claude relay
2026-03-30 19:37:21 +08:00
62b9aaa520
feat: prevent metadata from overriding model fields
2026-03-27 15:31:41 +08:00
0191a68d4e
Merge pull request #3400 from seefs001/fix/openai-usage
...
refactor: optimize billing flow for OpenAI-to-Anthropic convert
2026-03-23 15:03:57 +08:00
763c3ff709
Merge pull request #3331 from seefs001/fix/claude-beta-query
...
fix: apply forced beta query at final upstream URL stage
2026-03-23 15:03:36 +08:00
9ecad90652
refactor: optimize billing flow for OpenAI-to-Anthropic convert
2026-03-23 14:22:12 +08:00
2c3ae32c8e
fix map
2026-03-20 16:48:04 +08:00
498199b37d
fix code quality
2026-03-20 16:38:48 +08:00
ff29900f30
feat: Add support for counting cache-hit tokens in llama.cpp OpenAI-Compatible API
2026-03-20 16:10:18 +08:00
e520977efc
fix: apply forced beta query at final upstream URL stage
2026-03-19 15:49:50 +08:00
8aa8b81e03
fix: original_model && upstream_model paramOverrideKeyAuditPaths
2026-03-17 17:00:01 +08:00
bc80477b1a
feat: simplify param override audit UI and operation labels
2026-03-17 17:00:01 +08:00
5db25f47f1
feat: add param override audit modal for usage logs
2026-03-17 17:00:01 +08:00
da8a52f50a
feat: add conditional setting for HTTP headers in OpenRouter channel type
2026-03-12 19:05:30 +08:00
24ec27f844
Merge pull request #3221 from RedwindA/chore/updateModelList
...
chore: update model lists for frequently used channels
2026-03-12 15:13:03 +08:00
5e9cc681f5
feat: update header title for OpenRouter channel type
2026-03-12 15:05:58 +08:00
7e68e1b36a
chore: update model lists for frequently used channels
2026-03-11 23:39:18 +08:00
45a59d32fb
Merge pull request #3182 from seefs001/feature/params-override-beta-header-append
...
feat：support $keep_only_declared and deduped $append for header override
2026-03-10 02:03:02 +08:00
3b5d1daf39
Merge pull request #3148 from feitianbubu/pr/d8a25d36204224f8a4248b0ab3b03ba703796ea3
...
fix: kling risk fail return openAIVideo error
2026-03-09 22:19:04 +08:00
d087cc5025
feat：support $keep_only_declared and deduped $append for header token overrides
2026-03-09 00:12:53 +08:00
fb76abb329
fix: update OpenAI request fields to use json.RawMessage for dynamic data handling
2026-03-06 19:10:42 +08:00
2e20ede2a0
fix: kling risk fail return openAIVideo error
2026-03-06 16:32:52 +08:00
3defef3588
fix: ignore top_p
2026-03-06 12:07:36 +08:00
172f92aa72
fix: ignore top_p
2026-03-06 12:07:00 +08:00
12aacf27b6
fix: If top_p is not provided, Claude's logic will automatically set it to 1.
2026-03-06 12:03:51 +08:00
728607b8f5
Merge pull request #2769 from feitianbubu/pr/3d0aaa75866f8d958a777a7e7ac8c1e4b5b3e537
...
feat: kling cost quota support use FinalUnitDeduction as totalToken
2026-03-06 11:46:59 +08:00
cd1d43ae47
Merge pull request #3120 from nekohy/main
...
feats: repair the thinking of claude to openrouter convert
2026-03-05 18:10:46 +08:00
42500b3317
fix: fetch model add header passthrough rule key check
2026-03-05 17:49:36 +08:00
d2332685db
feat: add wildcard path support and improve param override templates/editor
2026-03-05 16:39:34 +08:00
a4629f2630
feats: repair the thinking of claude to openrouter convert
2026-03-05 06:12:48 +08:00
6f818574ab
fix: improve error message for unsupported image generation models
2026-03-04 00:36:03 +08:00
c79c1f95fd
fix: use default model price for radio price model
2026-03-03 11:22:04 +08:00
151264dfdc
Merge pull request #3081 from BenLampson/main
...
Return error when model price/ratio unset
2026-03-02 22:01:21 +08:00
618fce621b
Revert "fix: aws text content blocks must be non-empty"
2026-03-02 19:43:00 +08:00
Calcium-Ion