Model library
3
serverless models
1.05M
max context window
B200 · B300
Blackwell GPUs on demand
$5
free credits, no card
z.ai · LLM
Frontier agentic model. Top-tier coding, tool use, and long-context reasoning.
deepseek · LLM
Served on official DeepSeek V4.1-Flash weights — speed-tuned for high-volume chat, pipelines, and 1M-context work.
qwen · LLM
Balanced model for production chat — served on Qwen3.8-27B weights.
Pick a model, tune the knobs, copy the code or simulate a streaming reply.
Simulated streaming reply
Hit “Run demo stream” to preview token-by-token output…
from openai import OpenAI
client = OpenAI(
api_key="$RUNAII_API_KEY",
base_url="https://api.runaii.cloud/v1",
)
resp = client.chat.completions.create(
model="runaii/glm-5.3",
temperature=0.7,
top_p=0.9,
max_tokens=1024,
messages=[{"role": "user", "content": "Hello!"}],
)
print(resp.choices[0].message.content)~320ms
TTFT (typical)
218 tok/s
Peak throughput
1.05M
Context
180 req/min per API key after your first payment — same on every model and on both OpenAI and Anthropic dialects; 60 req/min while on welcome credits. Limits & 429 handling →
z.ai
Frontier agentic model. Top-tier coding, tool use, and long-context reasoning.
deepseek
Served on official DeepSeek V4.1-Flash weights — speed-tuned for high-volume chat, pipelines, and 1M-context work.
qwen
Balanced model for production chat — served on Qwen3.8-27B weights.
The advisor picks 3 open models sized to your use case, cost, and context needs — every price on this page.
Step 1 / 2
Open the playground with any model preloaded — $5 credit covers thousands of calls.