Summary
zai-glm-5-3 (Z.ai GLM 5.3) is listed as an active model on Mistral's public API but is missing from model_list.json. The predecessor zai-glm-5-2 is already in the catalog at the same pricing tier.
Evidence
Source 1 — Mistral models overview (docs.mistral.ai/getting-started/models):
Z.ai GLM 5.3 is listed as an active generalist model (open license, 1M-token context).
Source 2 — Dedicated model page (docs.mistral.ai/models/zai-glm-5-3):
Confirms API ID zai-glm-5-3, pricing ($1.40 input / $0.14 cached input / $4.40 output per 1M tokens), 1M context window, 128k max output. Status: Public Preview.
Existing analog
zai-glm-5-2 (line 11541 of model_list.json) uses the same pricing, context limits, and format. The new entry should mirror it.
Suggested spec
{
"zai-glm-5-3": {
"format": "openai",
"flavor": "chat",
"input_cost_per_mil_tokens": 1.4,
"output_cost_per_mil_tokens": 4.4,
"input_cache_read_cost_per_mil_tokens": 0.14,
"displayName": "Mistral: Z.ai GLM 5.3",
"reasoning": true,
"max_input_tokens": 1000000,
"max_output_tokens": 128000,
"available_providers": [
"mistral"
]
}
}
Summary
zai-glm-5-3(Z.ai GLM 5.3) is listed as an active model on Mistral's public API but is missing frommodel_list.json. The predecessorzai-glm-5-2is already in the catalog at the same pricing tier.Evidence
Source 1 — Mistral models overview (docs.mistral.ai/getting-started/models):
Z.ai GLM 5.3 is listed as an active generalist model (open license, 1M-token context).
Source 2 — Dedicated model page (docs.mistral.ai/models/zai-glm-5-3):
Confirms API ID
zai-glm-5-3, pricing ($1.40 input / $0.14 cached input / $4.40 output per 1M tokens), 1M context window, 128k max output. Status: Public Preview.Existing analog
zai-glm-5-2(line 11541 ofmodel_list.json) uses the same pricing, context limits, and format. The new entry should mirror it.Suggested spec
{ "zai-glm-5-3": { "format": "openai", "flavor": "chat", "input_cost_per_mil_tokens": 1.4, "output_cost_per_mil_tokens": 4.4, "input_cache_read_cost_per_mil_tokens": 0.14, "displayName": "Mistral: Z.ai GLM 5.3", "reasoning": true, "max_input_tokens": 1000000, "max_output_tokens": 128000, "available_providers": [ "mistral" ] } }