{"models":[{"id":"berget/zai-org/GLM-5.3-Flash","type":"llm","name":"GLM-5.3-Flash","model_id":"zai-org/GLM-5.3-Flash","capabilities":["text","tools","thinking","vision","reasoning","structured_output"],"speed":"fast","cost":0.059,"quality":"good","context_window":1048576,"max_output_tokens":131072,"description":"GLM-5.3-Flash is a natively multimodal model from Z.ai with 320B total / 18B active parameters. Its hybrid sparse-and-linear attention architecture keeps long-context behaviour accurate while cutting compute overhead, and it targets efficient coding and long-horizon agent tasks.","params":{"temperature":{"min":0,"max":2,"default":1},"reasoning":{"supported":["none","minimal","low","medium","high","xhigh","max"],"default":"","wire":"reasoning_effort"}},"pricing":{"billing_unit":"per_mtok","input":[0.279875],"output":[0.55975]},"region":"EU","country":"Sweden","model":"glm-5.3-flash","family":"glm-5.3-flash","pooled":true,"provider":{"slug":"berget","display_name":"Berget","owner_country":"SE"},"service_route":{"id":"berget/inference-eu","adapter":"berget"},"maker":{"slug":"zai","display_name":"Z.ai","country":"CN"},"compliance":{"residency":"EU","country":"Sweden","inference_location":"SE","zdr":{"logging":false,"moderation":false,"caching":true,"training":false,"subprocessors":false},"training":"no","logging":"none","moderation":"none","caching":"implicit","content_storage":"ephemeral","dpa_available":true,"third_party_access":null,"transfer_mechanism":"eu_resident","route_id":"berget/inference-eu"},"release":{"date":"2026-08-26","source":"vendor","url":"https://docs.z.ai/release-notes/new-released"},"quantization":"fp8","references":[{"kind":"model_card","url":"https://huggingface.co/zai-org/GLM-5.3-Flash","created_at":"2026-10-07T06:54:33.956988Z","updated_at":"2026-10-07T06:54:33.956988Z"},{"kind":"paper","url":"https://arxiv.org/abs/2602.15763","created_at":"2026-10-07T06:54:33.956988Z","updated_at":"2026-10-07T06:54:33.956988Z"}],"input_modalities":["text","image","video"],"output_modalities":["text"]}],"total":1,"offset":0,"limit":50}