鲨鱼魔法 is an AI platform offering access to official paid accounts for models such as Gemini T3, Vertex enterprise, Claude Max, and Volcano enterprise. It emphasizes stability, having handled billions of tokens without server crashes, and includes unique drawing models like Nano Banana (Pro) and NAI. The platform provides free cloud tavern usage upon registration, rebate channels for cost savings, and future game-based credit earning. It supports multiple endpoints for integration, including a chat completions URL and a base URL for tavern use. Technical support is available, and the platform complies with AI usage regulations for learning and research purposes.
openai.sharkmagic.top
鲨鱼魔法 (sharkmagic.top) provides access to official paid AI models like Gemini, Claude, and GPT through a unified proxy with OpenAI-compatible API.
- Models
- 1550 models
- From
- $9.00/M
- Speed
- 81 tok/s
- Updated
- 6/8/2026
- Latency
- 0.00 s
- Created At
- 1/11/2026
- Recharge Rate
- ¥0.01 per $1 quota
Features
DrawingTaskData Export
API Endpoints
- Tutorial Documentation
https://ucn3ufumwvhb.feishu.cn/wiki/SCqmwi5OriSL9RkxDhwc1pb0nxhView documentation
- Endpoint 2Historical / Unverified
https://ucn3ufumwvhb.feishu.cn - KOL Promotion Platform
https://kol.sharkmagic.com.cn/Earn referral rewards up to 25% commission
- Immortal Journey
https://ddct.sharkmagic.com.cn/AI game
- Endpoint 5Historical / Unverified
https://openai.sharkmagic.com.cn - Endpoint 6Historical / Unverified
https://kol.sharkmagic.com.cn - Endpoint 7Historical / Unverified
https://ddct.sharkmagic.com.cn - Endpoint 8Historical / Unverified
https://nai.sharkmagic.com.cn - Server Node
https://openai.sharkmagic.top/API main site
- Endpoint 10Historical / Unverified
https://openai.sharkmagic.top - Endpoint 11Historical / Unverified
https://openai.sharkmagic.online - Endpoint 12Historical / Unverified
https://st.sharkmagic.top - Endpoint 13Historical / Unverified
https://stol.sharkmagic.top
Claim this provider
Verify ownership to unlock provider management features:
- Edit provider name, content, and links
- Get featured with priority traffic and visibility boost
- Display a verified badge to build user trust
Leaderboard Rankings
Rankings are based on community-submitted tests and periodic health probes. Advisory only, not official data.
About 鲨鱼魔法
Health Check
100%Recent availability
History (72 pts)
PastNow
API Benchmarks & Pricing
Compare 4630 model rows across audit recency, latest speed tests, throughput, latency, and per-token pricing.
[lq]q|XQ/deepseek-v4-pro | - | - |
[lq]q|YS/gemini-3.1-pro-preview | - | - |
[mt]q|ccmax按次/claude-opus-4-5 | - | - |
[mt]q|ccmax按次/claude-opus-4-6 | - | - |
[ba]q|deepseek-ai/DeepSeek-V4-Flash-0731 | - | - |
[ba]q|按次/moonshotai/Kimi-K3 | - | - |
[ba]q|按次/zai-org/GLM-5.2 | - | - |
[ba]u|moonshotai/Kimi-K3 | - | - |
[ba]u|zai-org/GLM-5.2 | - | - |
[ba]u|zai-org/GLM-5.3 | - | - |
[ba]u|zai-org/GLM-5.3-Flash | - | - |
[ff]u|gemini-2.5-flash-image-banana2 | - | - |
[ff]u|gemini-3-pro-image-preview-banana2 | - | - |
[ff]u|gemini-3.1-flash-image-preview-banana2 | - | - |
[free]DeepSeek-V3.2tier_3 | Free | Free |
[free]DeepSeek-V3.2tier_2 | Free | Free |
[free]DeepSeek-V4-Flashtier_2 | Free | Free |
[free]DeepSeek-V4-Flashtier_3 | Free | Free |
[free]DeepSeek-V4-Flash-0731 | - | - |
[hm]q|ais/gemini-3.7-flash | - | - |
[hm]q|ais/gemini-3.8-flash | - | - |
[hm]q|anti/gemini-3.8-flash | - | - |
[hm]q|build/gemini-3.7-flash | - | - |
[hm]q|build/gemini-3.8-flash | - | - |
[hm]q|deepseek-v4-pro-0813-c | - | - |
Showing 25 of 4630 model rows
Recent Test Records
| Time | Model | Speed | Latency |
|---|---|---|---|
| Sep 3, 02:37 PM | [hm]t|deepseek-v4-pro-0813 | 144.07 tok/s | 13.51s |
| Sep 3, 02:36 PM | [hm]t|deepseek-v4-pro-0813 | 115.71 tok/s | 16.98s |
| Dec 15, 12:49 PM | deepseek-r1 | 56.00 tok/s | 1.50s |
| Aug 12, 02:48 PM | grok-3 | 28.05 tok/s | 6.44s |
| Aug 12, 02:47 PM | claude-4-sonnet | 50.48 tok/s | 3.47s |
| Aug 12, 02:25 PM | [503]gemini-flash-lite | 229.64 tok/s | 2.17s |
| Aug 12, 02:23 PM | [503]gemini-flash | 199.70 tok/s | 10.04s |
| Aug 12, 02:20 PM | [503]gemini-pro | 82.99 tok/s | 22.12s |
| Jul 23, 04:20 PM | grok-3 | 23.68 tok/s | 1.58s |
| Jun 16, 05:39 AM | claude-4-sonnet | 31.73 tok/s | 7.93s |
Similar API Provider Alternatives to Compare
Compare 鲨鱼魔法 alternatives against 6 nearby API providers using 544 LMSpeed signals across shared model coverage, pricing, benchmark speed, uptime, and free-model availability.
| Provider | Why compare | Models | Free | Avg price | Speed | 30d uptime |
|---|---|---|---|---|---|---|
| 鲨鱼魔法 openai-sharkmagic-top 鲨鱼魔法 (sharkmagic.top) provides access to official paid AI models like Gemini, Claude, and GPT through a unified proxy with OpenAI-compatible API. | Current provider baseline | 1,550 | 2 | $9.00/M | 81 tok/s | 9970% |
| V-API v-api Provides access to over 500 AI models via OpenAI protocol with pay-as-you-go pricing, fast response times, and transparent billing. |
| 699 | 15 | $0.010/M | 130 tok/s | 9950% |
| Elysiver API elysiver-api Elysiver API appears to provide an OpenAI-compatible API gateway at elysiver.h-e.top. The public root is protected by a Cloudflare challenge during review. |
| 78 | 0 | $0.014/M | 119 tok/s | 9960% |
| ABC Relay www-abcrelay-com A unified API gateway providing standardized access to multiple large language and AI models from various providers. |
| 54 | 0 | $0.029/M | 53 tok/s | 9980% |
| 丸美小沐写作 wanmei-xiaomu-xiezuo A Windows desktop application for AI-assisted novel writing, supporting multiple AI models and batch generation. |
| 28 | 22 | N/A | 147 tok/s | 9960% |
| 哈基米API api-123chat-top 哈基米API provides an OpenAI-compatible API relay service for accessing multiple AI models including GPT, Claude, and more. |
| 20 | 6 | N/A | 175 tok/s | 0% |
| Koyeb AI Gateway new-api-koyeb-app An OpenAI-compatible API gateway deployed on Koyeb, providing access to multiple AI models. |
| 578 | 420 | N/A | 33 tok/s | 9950% |
Announcements
default5/15/2026
Xianyu has been banned. If you have any issues, please contact after-sales group customer service.
default12/19/2025
Platform sign-in system in beta, daily sign-in to earn points, points can be exchanged for various rewards, come check it out in the after-sales group on the homepage~
success12/11/2025
Due to force majeure, the platform will switch to a new address. This address (with .com.cn suffix. If you entered from .top suffix, ignore this announcement.) will be discontinued in 7 days. Please add the after-sales group promptly to get the latest address.
Old server shutdown completed.
default11/26/2025
The platform is distributing Cloud Tavern for free, come claim it in the after-sales group~
default11/21/2025
Starting now, anyone who has accumulated a certain amount of consumption on the platform can claim a free Cloud Tavern monthly card in the after-sales group, chat anytime, anywhere, click on the after-sales group on the homepage to enjoy~
warning11/8/2025
Recently, Google's risk control is strict. If you don't know which model is available, you can check in the after-sales group. The after-sales group number is on the platform homepage or in the announcements.
warning10/14/2025
Recently, due to Google restrictions, model fluctuations are significant. For the specific status of models, please check the after-sales group. Thank you~
After-sales group is on the homepage or in the announcements.
error9/14/2025
The practice arena currently has a critical bug. It should have been closed directly, but due to feedback from some users who have this need, it has been reopened. This bug will cause calls in the practice arena to deduct not only the remaining quota but also the quota limit (with no logs). Please do not continue using the practice arena. If you continue to use it, you are deemed to agree to bear the consequences, and we will not provide compensation. Thank you for your understanding.
default9/12/2025
Weekly maintenance announcement: Dear users, the platform will begin weekly update maintenance on September 14th at 10:00 AM. The estimated downtime is 1 hour. Please try to avoid sensitive operations during this time. Thank you for your support~
default9/6/2025
Please add the after-sales group mentioned in the announcements or on the homepage promptly to avoid missing important notices. Future announcements will only be notified within the group.
ongoing9/5/2025
Weekly maintenance announcement: Dear users, the platform will begin weekly update maintenance on September 6th at 10:00 AM. The estimated downtime is 1 hour. Please try to avoid sensitive operations during this time. Thank you for your support~
default8/27/2025
Models without prefix changed to official self, just modify the model name directly.
success8/26/2025
Limited-time event launched! Volcano Doubao models, all 50% off! Valid until September 4th (excluding 1.6-think model)
default8/22/2025
Weekly maintenance announcement The platform will begin weekly update maintenance on August 24th at 10:00 AM. The estimated maintenance time is 30 minutes. Please try to avoid sensitive operations during this time. Thank you for your support~
default8/20/2025
Platform update, Volcano official models, all 30% off except v3!!! Official channel quality guaranteed.
ongoing8/20/2025
Hello users, in order to unify the platform's model naming conventions, the platform will standardize model naming tonight. You may need to re-copy and configure the interface modelId to continue using it. Thank you for your long-term support.
default8/17/2025
The prices of most models on the platform will be kept as stable as possible, but occasionally adjustments may be needed due to channel or official model price changes. Notifications will be given in the after-sales group at that time. Therefore, please add the after-sales group promptly for better service. Thank you.
warning8/16/2025
Based on statistical analysis, the estimated availability time for the gemini-2.5-pro model is: Beijing Time (UTC+8): 2:00 - 15:00, Coordinated Universal Time (UTC): 18:00 - 7:00. (Statistical data may have errors, please refer to actual conditions.)
success8/15/2025
Weekly maintenance announcement The platform will begin weekly update maintenance on August 16th at 10:00 AM. The estimated maintenance time is 30 minutes. Please try to avoid sensitive operations during this time. Thank you for your support~
Completed
Notes
- Health checks: Scope: the 72-hour chart and recent availability measure API connectivity only. Each bar summarizes one hour of checks. Targets: LMSpeed tries the configured health check URL and provider status URL first, then API endpoints derived from known API hosts and recent speed-test base URLs. A website host is considered only when it looks like an API endpoint. Probe steps: each candidate goes through DNS lookup, TCP connection, TLS handshake for HTTPS, and an HTTP HEAD request with redirects followed. Probing stops after the first reachable candidate. Reachable criteria: every required network step must succeed. An HTTP response below 500 is treated as reachable, including 401 because it confirms that an authenticated API endpoint responded, except for statuses classified as blocked. Blocked results: HTTP 403, 429, 521, 525, and 530, plus detected WAF or Cloudflare challenges, are shown as blocked and excluded from availability calculations because LMSpeed cannot determine whether the API itself is down. Model availability: when a dedicated test key is configured, LMSpeed sends an authenticated GET request to a derived /models endpoint and compares returned model IDs with this provider's listed models. These per-model results appear in Models & Pricing and are not included in the provider connectivity percentage. Timeouts: TCP connection, TLS handshake, HTTP connectivity, and model requests each use a 20-second timeout. A full run can take longer when several candidates are tried. Frequency: a background worker checks all providers every 5 minutes by default. The 72-hour chart combines those samples into hourly bars, and the schedule may be changed by the service operator. Limit: automated samples are not an SLA and do not guarantee account quota, every model, every region, or successful completion requests. Check the provider's own status page before making operational decisions.
- Domain Rating data is sourced from Ahrefs. It is a 0–100 backlink-based domain strength signal and does not measure API speed or reliability.
- Announcements and FAQ are read from this provider's NewAPI status snapshot when available. LMSpeed stores the original content and optional English translations from the provider status source, then shows the localized fields on this page.