Apertis AI

06/17/2026

✨ The latest GLM 5.2 is now live on Apertis!
GLM-5.2 is Z.AI's flagship model for long-horizon task ex*****on, designed to handle complex, project-scale workflows with high reliability. Featuring a 1M-token context window, it can maintain and reason over extensive engineering context, enabling consistent ex*****on across large, multi-stage tasks.
Enjoy it!
Try it now on Apertis: https://apertis.ai/models/glm-5.2

GLM 5.2 by Z.AI — Access via Apertis API. Input: $0.0014/1M tokens, Output: $0.0044/1M tokens. Context: 1.0M tokens. OpenAI-compatible API with 50% cost…

06/12/2026

✨ The latest Kimi K2.7 Code is now live on Apertis!
Kimi K2.7 Code is a coding-focused model in Moonshot AI's Kimi K2 family, designed for long-horizon software engineering and agentic development workflows. Built on a native multimodal Mixture-of-Experts (MoE) architecture, it supports text, image, and video inputs and operates exclusively in thinking mode, preserving reasoning across multi-turn interactions.
Enjoy it!
Try it now on Apertis:

Kimi K2.7 Code by Moonshot AI — Access via Apertis API. Input: $0.0009/1M tokens, Output: $0.0040/1M tokens. Context: 262K tokens. OpenAI-compatible API…

06/10/2026

✨ Claude Fable 5 model is on Apertis Chat
Claude Fable 5 is Anthropic's Mythos-class model, designed for autonomous knowledge work, coding, and long-running agentic workflows.

Access with Apertis: https://apertis.ai

06/10/2026

✨ The latest Claude Fable 5 is now live on Apertis!
Claude Fable 5 is Anthropic's Mythos-class model, designed for autonomous knowledge work, coding, and long-running agentic workflows. It supports text, image, and file inputs with text output, includes reasoning capabilities, and features a 1M-token context window for handling complex, high-context tasks.
Enjoy it!
Try it now on Apertis:

Claude Fable 5 by Anthropic — Access via Apertis API. Input: $0.0100/1M tokens, Output: $0.0500/1M tokens. Context: 1.0M tokens. OpenAI-compatible API…

06/09/2026

Picking an AI model — are you guessing from the pricing table, or have you actually measured it?

/cli lets you benchmark models head-to-head from your terminal in one command.

🔹 Choose before you ship — run the same prompt across models and see which is fastest and cheapest
🔹 Real cost — measured from your usage delta, not a pricing-table estimate
🔹 Performance data — client + server-side latency and tokens/sec, all in one comparison
🔹 Track spend anytime — one command shows your key's lifetime usage and cost
🔹 CI-friendly — every command supports --json to pipe straight into scripts or dashboards

📚 Docs: docs.apertis.ai/api/sdks/cli
📦 npm: https://npmjs.com/package//cli → npm i -g /cli

06/09/2026

Building AI agents on Apertis?
The two things that bite hardest: loops that quietly burn your budget, and risky actions with no one in the loop. /agent handles both — you define the tools and when to stop, it runs the whole loop for you.

🔹 Coding / research agents — let the model chain multiple tools (read files, search, compute) and converge on an answer
🔹 Hard budget cap — maxCost stops on real measured spend, not an estimate, so a runaway loop can't drain your balance
🔹 Human in the loop — flag tools like delete, refund, or send-email for approval; the loop pauses until you say go
🔹 Live output — stream text and reasoning to your UI as the agent works
🔹 Easy to adopt — used a similar agent SDK before? The API is nearly identical; most code ports with a one-import swap

📚 Docs: https://docs.apertis.ai/api/sdks/agent-sdk
📦 npm: https://npmjs.com/package//agent

06/09/2026

✨ New in Apertis Chat: multi-model conversations at once.

Ask once. Compare multiple AI models side by side in the same chat. Spot blind spots, cross-check reasoning, and make faster, more confident decisions without switching tools.

Great teams do not rely on one answer. Now your chat does not have to either.

Try it now: compare GPT-5.5, Claude Opus 4.8, DeepSeek V4 Pro, and 470+ models in Apertis Chat.

06/06/2026

✨ Redesigned Chat Interface
As fine-grained redesign UIUX on Apertis Chat interface, the updated version provides:
1. More precise information delivery
2. Bugs resolved
3. Brand new UIUX follows latest Apertis branding

🔥 Coming: Compare multiple models responses in one session.

Enjoy it.
https://chat.apertis.ai

06/05/2026

✨ The latest Nemotron 3 Ultra is now live on Apertis!
NVIDIA Nemotron 3 Ultra is an open frontier reasoning and orchestration model featuring a 550B-parameter Mixture-of-Experts (MoE) architecture with 55B active parameters per token. Built on a hybrid Transformer–Mamba design, it supports text input and output with a 1M-token context window, enabling large-scale reasoning and long-horizon task ex*****on.
Enjoy it!
Try it now on Apertis:

Nemotron 3 Ultra by NVIDIA — Access via Apertis API. Input: $0.0005/1M tokens, Output: $0.0025/1M tokens. Context: 1.0M tokens. OpenAI-compatible API with…

06/04/2026

✨ The latest Qwen3.7 Plus is now live on Apertis!
Qwen3.7-Plus is a cost-effective multimodal model in Alibaba's Qwen3.7 series, supporting text and image inputs with text output. It combines the series' strong language capabilities with significantly enhanced vision-language understanding, while retaining full-stack agent-level intelligence for coding, tool use, and productivity workflows.
Enjoy it!
Try it now on Apertis: https://apertis.ai/models/qwen3.7-plus