Compute Token operates as an API gateway, offering a unified interface to access various large language models. It supports models from providers including OpenAI (GPT-4o), Anthropic (Claude), Google (Gemini), DeepSeek, Qwen, xAI (Grok), Moonshot AI, Zhipu AI, Volcengine, Cohere, Minimax, Baidu (Wenxin), iFlytek (Spark), and Tencent (Hunyuan). The service also lists support for Azure AI, Xinference, Suno, and Midjourney. It provides a standard OpenAI-compatible API endpoint (https://computetoken.ai/v1), allowing developers to integrate with a single SDK. The platform is marketed with a focus on high-speed, stable connectivity and continuous addition of new models.
Compute Token
computetoken.ai
A unified API gateway providing access to multiple large language models through a single endpoint.
- Models
- 64 models
- From
- $0.0096/M
- Speed
- 22 tok/s
- Updated
- 6/8/2026
- Latency
- 0.00 s
- Created At
- 4/23/2026
- Recharge Rate
- ¥7.00 per $1 quota
Features
Login Methods
API Endpoints
- Endpoint 1Historical / Unverified
https://computetoken.ai
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 Compute Token
Health Check
API Benchmarks & Pricing
Compare 80 model rows across audit recency, latest speed tests, throughput, latency, and per-token pricing.
moonshotai/kimi-k3 | - | - |
grok-4.20-0309-non-reasoningxai | $0.012/M | $0.024/M |
grok-4.20-0309-reasoningxai | $0.012/M | $0.024/M |
grok-composer-2.5-fastxai | $0.719/M | $0.719/M |
gpt-5.6-lunacodex | $0.288/M | $1.73/M |
gpt-5.6-terracodex | $0.719/M | $4.32/M |
gpt-5.6-solcodex | $1.44/M | $8.63/M |
grok-4.5xai | $0.719/M | $0.719/M |
claude-sonnet-5ccmax | $0.767/M | $3.84/M |
claude-sonnet-5claude_official | $1.15/M | $5.75/M |
gemini-3.1-flash-image-previewgemini_official | $0.336/M | $40.27/M |
gemini-3-pro-image-previewgemini_official | $1.34/M | $80.55/M |
glm-5.2default | $0.336/M | $1.05/M |
claude-fable-5ccmax | $3.84/M | $19.18/M |
claude-fable-5claude_official | $5.75/M | $28.77/M |
claude-opus-4-8claude | $0.959/M | $4.79/M |
claude-opus-4-8default | $1.20/M | $5.99/M |
claude-opus-4-8ccmax | $1.92/M | $9.59/M |
claude-opus-4-8claude_official | $2.88/M | $14.38/M |
| $0.0096/M | $0.019/M | |
claude-opus-4-7claude | $0.959/M | $4.79/M |
claude-opus-4-7default | $1.20/M | $5.99/M |
claude-opus-4-7ccmax | $1.92/M | $9.59/M |
claude-opus-4-7claude_official | $2.88/M | $14.38/M |
grok-4.3xai | $0.012/M | $0.024/M |
Showing 25 of 80 model rows
Recent Test Records
| Time | Model | Speed | Latency |
|---|---|---|---|
| Apr 23, 10:17 AM | gpt-5.3-codex | 21.50 tok/s | 3.14s |
Similar API Provider Alternatives to Compare
Compare Compute Token alternatives against 6 nearby API providers using 361 LMSpeed signals across shared model coverage, pricing, benchmark speed, uptime, and free-model availability.
| Provider | Why compare | Models | Free | Avg price | Speed | 30d uptime |
|---|---|---|---|---|---|---|
| Compute Token computetoken-ai A unified API gateway providing access to multiple large language models through a single endpoint. | Current provider baseline | 64 | 0 | $0.014/M | 22 tok/s | 9980% |
| 钱多多 API api2-aigcbest-top Provides AI-generated content APIs for various applications, including text and image generation. |
| 1,418 | 2 | $0.0025/M | 101 tok/s | 9940% |
| TommyLam API new-api-tommylam-me TommyLam API offers an OpenAI-compatible API gateway for accessing multiple AI models at competitive rates. |
| 231 | 100 | N/A | 61 tok/s | 9990% |
| F2API api-f2api-com F2API is an enterprise AI gateway providing a unified OpenAI-compatible interface for accessing multiple large language models. |
| 1,262 | 0 | N/A | 272 tok/s | 9990% |
| 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 | 9960% |
| Yun API api-zyai-online Yun API is a high-speed, stable, and high-concurrency API relay service based on official APIs, offering access to over 100 AI models including text, image, and video generation. |
| 431 | 0 | $0.00000342/M | 32 tok/s | 9940% |
| 哈基米API站 api-gemai-cc 哈基米API站 offers a unified API gateway for large language models with competitive pricing and stable service. |
| 323 | 0 | $0.930/M | 119 tok/s | 9990% |
Announcements
Codex group usage rate lowered to 0.1; the account pool is the pro account pool. Stable 24/7 all year round. If you have any questions during use, please contact the customer service email listed in the announcement on the homepage for feedback!
【Model Price Adjustment Notice】
claude-opus-5 has been added. Meanwhile, prices for some GPT models have been reduced:
gpt-5.6-luna overall price cut by 80% gpt-5.6-terra has also been reduced; visit the site for details
The 7/14x bonus recharge promotion has ended, and the platform will uniformly process historical promotions. From the date of this announcement, if any promotional recharge balance remains, the platform will refund the actual paid amount of the corresponding order via the original payment method (the used portion will be treated as a promotional bonus). Accounts that have fully used their promotional balance and regular recharge balances are unaffected. No application is needed for refunds; the platform will process them uniformly. Crediting time depends on the actual processing progress of the payment channel. In addition, the check-in bonus balance activity has been discontinued; existing balances can be used until May 31, 2026, after which unused portions will expire. This announcement was published on May 11, 2026.
Regarding Refunds
Dear all,
We have to post this and apologize first.
Recently, A社's risk control has been very strict. We've tried multiple layers of proxies, constantly changing IP pools, and updating channels, but this time A社 is doing a comprehensive sweep that we can't withstand.
So, we have to pause cc services. All remaining balances for recharged users will be fully refunded.
Don't worry; refunds will be processed gradually.
If we find a more stable solution, we'll notify everyone. Previous promises (stable, user-friendly, cost-effective) remain.
These days have been turbulent—sometimes stable, then blocked again. In the end, we couldn't hold on.
We're sorry for the inconvenience.
We remember all supporters. In the future, we hope to give back more.
Refunds will be automatically returned to the original method, processed ASAP.
That's all.
Apologies again!!
CC channel triggered risk control.
CC channel is back to normal!
Regarding Claude Channel Issues
Hello everyone, first, we sincerely apologize.
Yesterday was our first day online, and the cc channel had some issues affecting normal use. Especially for friends who have recharged and started using it, we're really sorry; we take responsibility for this. We are a small team, and encountering this on the first day of launch makes us more anxious than anyone. We are now urgently investigating and maintaining; fixing what can be fixed, and not leaving it unattended.
Users who have recharged don't need to worry; we will be responsible for your rights, and we will sync progress as soon as it's restored.
For those who are stockpiling or observing, we want to say: You can continue to come; our team is still here, and the site is still up. The first-day crash isn't glorious, but we don't avoid it: when there's a problem, we acknowledge it, fix it, and fix it until it's good. Private messages are too many to reply to one by one; everyone is welcome to give feedback in the group, and we will read and handle it. Do first, talk later, see the results! Thank you for your understanding!
We will update as soon as there's progress.
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.

