GLM 5.2 and GLM 5.3 were retired: free alternatives through the OpenAI API

GLM 5.3 was retired by NVIDIA on 2026-10-28 at 08:00 UTC. A direct call now returns 410 Gone: "The model 'z-ai/glm-5.3' has reached its end of life". GLM 5.2 was retired in August 2026. No spelling of either reaches a live model.

GLM 5.3 Flash (z-ai/glm-5.3-flash) is still listed, but it is slow: on 2026-10-28 it took 59 seconds to answer "Say ok.", and with X-Inferall-Routing: exact it timed out at 180 seconds 4 of 4 times on 2026-10-27.

What happens if you still send z-ai/glm-5.3

• Default routing: InferAll answers with another free model and tells you which one: response.model and the X-Inferall-Actual-Model header name it. On 2026-10-28 that was Nemotron 3 Ultra or gpt-oss-20b.
• X-Inferall-Routing: exact: the call fails with HTTP 410, because exact routing never substitutes a model.

What to send instead

nvidia/nemotron-3-ultra-550b-a55b is the largest free model we serve and the default (guide). It answered a 210,000-token prompt on 2026-10-28, so it also suits long-context work. The live list is at GET /ai/v1/models.

Python (OpenAI SDK)

from openai import OpenAI

client = OpenAI(
    base_url="https://api.inferall.ai/v1",
    api_key="ifu_your_key_here",  # inferall.ai/keys
)

response = client.chat.completions.create(
    model="nvidia/nemotron-3-ultra-550b-a55b",
    messages=[{"role": "user", "content": "Write a Python one-liner that reverses the words in a sentence."}],
    max_tokens=600,
)
print(response.model)  # the model that answered
print(response.choices[0].message.content)

Claude Code

export ANTHROPIC_BASE_URL=https://api.inferall.ai
export ANTHROPIC_API_KEY=ifu_your_key_here
export ANTHROPIC_MODEL=nvidia/nemotron-3-ultra-550b-a55b
claude

History: GLM 5.3 before the retirement

What we measured through InferAll on a one-line coding prompt, for the record:

How it was called	Who answered	Time per answer
z-ai/glm-5.3, default routing (2026-10-19)	Nemotron 3 Ultra, 5 of 5	16s to 23s
z-ai/glm-5.3, exact routing (2026-10-19)	GLM 5.3, 5 of 5	48s to 134s
z-ai/glm-5.3, exact, reasoning_effort: "low" (2026-10-27)	GLM 5.3, 5 of 5	36s to 60s

Get a key

InferAll is an LLM aggregator: one API key for free open models and paid OpenAI, Anthropic, Google and xAI models, through OpenAI- and Anthropic-compatible endpoints.

Sign up at inferall.ai/keys. New accounts get 25 free calls on open models, no card needed.