💥 Price cut — Kimi K3 now 42% below official ($1.73 / $8.65 vs $3 / $15) · Qwen 3.8 Max 48% below ($1.04 / $3.11) · first top-up +50% bonus
Up to 50% below international pricing · Card payments live · Free credit on signup

Your tools, powered by
China's frontier models

One key for the newest flagships from all four of China's frontier labs — Qwen 3.8 Max, Kimi K3, DeepSeek V4 and GLM 5.2 — the official, full-quality models, not quantized community mirrors, at up to 48% below the vendors' international rates — cached tokens included. Native to Claude Code, Cline and Cursor in two lines of config.

OpenAI & Anthropic compatible Low latency across APAC Every request itemized
terminal — Claude Code on qwen3.8-max
# two lines. that's the whole migration.
export ANTHROPIC_BASE_URL="https://api.gloritoken.ai"
export ANTHROPIC_AUTH_TOKEN="sk-your-key"

$ claude
✔ Connected · model: qwen3.8-max
> refactor this module and add unit tests…
4+frontier Chinese models, one key
1Mcontext window · whole-repo aware
5–10×cheaper than US flagship models
100%itemized, auditable billing
THE MODELS

The July 2026 flagship lineup — one key

Qwen 3.8 · Kimi K3 · DeepSeek V4 · GLM 5.2 — the newest flagship from each of China's four frontier labs, first-hand from the makers' official APIs, full-quality builds. Priced up to 48% below the vendors' international rates — Qwen 3.8 Max at 48% below, Kimi K3 freshly cut to 42% below, and cached input discounted at the same ratio.

Featured flagships official APIs · full-quality · no quantized mirrors
QWEN · NEW FLAGSHIP
qwen3.8-max48% OFF INTL
Alibaba's 2.4T-parameter MoE flagship — multi-day autonomous coding runs, native vision across plan→execute→verify. $1.04 / $3.11 vs $2 / $6 international.
$1.04in /1M
$3.11out /1M
1M ctx
MOONSHOT · PRICE CUT
kimi-k3NOW 42% OFF OFFICIAL
Moonshot's 2.8T flagship — 1M context, native vision. Cut from $2.40/$12.00 to $1.73/$8.65 — vs $3/$15 official.
$1.73in /1M
$8.65out /1M
1M ctx
ZHIPU
glm-5.247% OFF INTL
Zhipu's newest — strong at both coding and general tasks, 1M context.
$0.74in /1M
$2.57out /1M
1M ctx
QWEN
qwen3.7-plusbest value
Multimodal flagship — text, image & video input at a mid-tier price.
$0.18in /1M
$0.74out /1M
256K ctx
DEEPSEEK
deepseek-v4-provendor parity
The all-round national flagship — served at DeepSeek's own international rate. We earn nothing here; it exists so one key covers every lab.
$0.435in /1M
$0.87out /1M
1M ctx
DEEPSEEK
deepseek-v4-flashhigh volume
Extreme price-performance — completions, high-volume and batch jobs.
$0.09in /1M
$0.18out /1M
1M ctx
More models — coding, long-context & exclusives same key · pick per request by model name
QWEN · NEW
qwen3.7-flashmultimodal · value
Vision-language Flash — multimodal agents and high-frequency calls at rock-bottom cost.
$0.02in /1M
$0.07out /1M
1M ctx
MINIMAX · NEW
minimax-m3coding · agentic
MiniMax's flagship — strong coding & agentic performance, native multimodal, 1M context.
$0.39in /1M
$1.54out /1M
1M ctx
MOONSHOT · CODING
kimi-k2.7-codeagentic coding
Moonshot's dedicated coder — reliable instruction-following across long context.
$0.60in /1M
$2.48out /1M
256K ctx
MOONSHOT · EXCLUSIVE
kimi-k2.7-code-highspeedonly here
High-speed Kimi coder (~180–260 tok/s) — no international provider offers it.
$1.89in /1M
$7.85out /1M
256K ctx
ZHIPU · HIGH SPEED
glm-5.2-fast-preview1.5–2× faster
Speed-tuned GLM 5.2 — same capability as the standard build at 1.5–2× the output throughput. For live chat, multi-turn agents and streaming code gen. 1M context.
$1.47in /1M
$5.15out /1M
1M ctx
QWEN · LEGACY
qwen3.7-max12% OFF INTL
Previous-gen flagship (snapshot 2026-05-20) — for workflows pinned to Qwen 3.7. New work should use qwen3.8-max.
$1.10in /1M
$3.31out /1M
1M ctx

* Pay-as-you-go, USD per 1M tokens. Alibaba-served models are 12–48% below the vendors' international list rates, and cached input is discounted at the same ratio — the cache maths works exactly as it does going direct. Two official-direct models (deepseek-v4-pro, kimi-k2.7-code-highspeed) are at vendor parity; we earn nothing on them. Full breakdown, including who should not use us. Models run on the makers' official APIs (mainland-China endpoints) via our Singapore gateway; we don't store prompt contents and never train on your data. We do not relay Claude, GPT or Gemini.

PRICING & PAYMENT

Start free. Top up when you need to.

No subscription, no minimum commitment — you pay only for the tokens you use, at up to 48% below the model makers' international rates.

Start free

$0
Sign up, verify your email, get free credit instantly — no card required
  • Your free credit goes further than you think — roughly 5.5M output tokens on deepseek-v4-flash
  • Every model, pay-as-you-go
  • Usage dashboard, itemized per-request billing
  • Works with Claude Code, Cline, Cursor and any OpenAI-compatible tool
Get your API key

Top up

$10 · $20 · $50
Credit never expires. Balance stops at zero — you can never be overbilled.
  • First top-up gets +50% bonus credit — limited-time launch offer
  • Pay by card — secure Stripe checkout in USD
  • Credited to your balance the same day, usually within minutes
  • Your API key never changes
  • Invoice and volume pricing available on request
Top up by card

* Prepaid credit, billed in USD. Consumed usage is non-refundable; any unused balance can be refunded on request within 30 days — see Terms. Questions about billing: hello@gloritoken.ai.

WORKS WITH YOUR TOOLS

Keep your workflow. Swap the engine.

Anything that speaks the OpenAI or Anthropic API works out of the box — point it at gloritoken and you're running on China's frontier coding models.

Claude CodeNative Anthropic API

Qwen CodeOpenAI protocol

ClineVS Code extension

CursorCustom base URL

Cx

Codex CLICustom provider

RC

Roo CodeOpenAI-compatible

OpenCodeCustom models

CS

Cherry StudioDesktop client

Cb

ChatboxDesktop client

+

Any toolOpenAI-compatible

* Tool names and logos belong to their respective owners; shown for compatibility only, no affiliation or endorsement implied.

FAQ

Fair questions

Where does my code actually go?

Our gateway runs in Singapore; inference happens on the model makers' official APIs, which are served from mainland China — we state that plainly rather than hide it. We don't store your prompt contents and never train on your data; we keep only the usage metadata needed for billing.

How is this different from OpenRouter?

OpenRouter routes you to whichever provider is cheapest — often a quantized, community-hosted copy that varies in quality and uptime. gloritoken serves the official, full-quality models first-hand, natively over the Anthropic protocol (real Claude Code support), with Alibaba-served models priced 12–48% below the vendors' international rates — and one key covers every lab.

Does it really work with Claude Code?

Yes — natively. We expose a real Anthropic-protocol endpoint (/v1/messages), so Claude Code connects with two environment variables, no proxy hacks. Cline, Roo Code, Cursor and any OpenAI-compatible tool work the same way through /v1.

Swap your engine today

The API is live and self-serve: sign up, verify your email, and start with free credit — instantly, no waiting. Need a bigger trial or a team plan? Tell us what you're building.