# GLM-5.3: official model and deployment card

The full GLM-5.3 model card, with serving routes and evaluation notes. Keep it separate from Flash and from quantized community conversions.

- Type: Official model card
- Evidence: source_reported; site reproduced: false
- Model family: GLM
- Model: zai-org/GLM-5.3
- Original author: zai-org
- Original project: https://huggingface.co/zai-org/GLM-5.3
- Pinned document: https://huggingface.co/zai-org/GLM-5.3/blob/ba32f8c19c8738e2c57c71798c601d477a7bf844/README.md
- Source revision: ba32f8c19c8738e2c57c71798c601d477a7bf844
- Source commit time (not experiment date): 2026-10-14T12:22:48.000Z
- Checked: 2026-10-17
- Experiment date: None
- Runtime topics: vLLM, SGLang, Transformers, KTransformers, TokenSpeed
- Hardware topics: 
- Reported setup: {"system": null, "node_count": null, "parallelism": null, "kv_dtype": null, "context_limit_tokens": null, "experiment_date": null}
- Performance figures: attributed selections with per-point conditions; no cross-source ranking

## Conditions

Unknown is null. Recipe options are not measured benchmark settings.

```json
{
  "conditions": {
    "model_revision": "ba32f8c19c8738e2c57c71798c601d477a7bf844",
    "weight_hash": null,
    "runtime_execution_revision": null,
    "input_tokens": null,
    "output_tokens": null,
    "concurrency": null,
    "cache_state": null,
    "quantization": null,
    "speculative_decoding": null,
    "temperature": null,
    "top_p": null
  },
  "test_profiles": [],
  "recipe_parameters": null,
  "deployment_options": []
}
```

## Reading notes

- Reasoning effort and conversation handling are documented with the checkpoint.
- The card links a separate Ascend deployment route.

## Limits

- Published capability evaluations do not demonstrate latency on the reader’s hardware.

## License declaration

{"declared_spdx": "other", "declared_name": "glm-5.3", "scope": "Publisher model-card declaration; inspect the linked LICENSE and artifact-specific terms."}


## Editorial answer

The full GLM-5.3 model card, with serving routes and evaluation notes. Keep it separate from Flash and from quantized community conversions.

By Run AI Here editorial · 2026-10-17

AI-assisted source reading and original editorial synthesis; no independent model execution.

## Which artifact does this page identify?

This is the full GLM-5.3 card, separate from Flash and from community quantizations. Its local-serving section includes a distinct Ascend route. Use the exact checkpoint and the selected conversion as separate identities. The companion recipe’s 743B total and 39B active labels belong to that recipe rather than a measured memory receipt.

Sources: https://huggingface.co/zai-org/GLM-5.3/blob/ba32f8c19c8738e2c57c71798c601d477a7bf844/README.md#L41

## What is the correct reasoning configuration?

The publisher’s controls match the documented low, high and max reasoning-effort levels, defaulting to max. clear_thinking remains a separate chat-template option with a false default. Retaining the default effort matters for benchmark reproduction; enabling clear_thinking for a chat workflow should be recorded as its own configuration choice.

Sources: https://huggingface.co/zai-org/GLM-5.3/blob/ba32f8c19c8738e2c57c71798c601d477a7bf844/README.md#L55

## How much of the published score can be reproduced from a headline?

The card reports 66.9 on DeepSWE v1.1 and 88.2 on Terminal-Bench 2.1. Their footnotes use different sampling, context and timeout settings. Terminal-Bench 3.0 is another protocol: three rollouts per task, isolated task containers, up to 600 turns and a ten-hour timeout. A score from one task is not a prediction of workstation throughput or another task’s accuracy.

Sources: https://huggingface.co/zai-org/GLM-5.3/blob/ba32f8c19c8738e2c57c71798c601d477a7bf844/README.md#L26

## What should be retained before comparing results?

Keep the exact task revision, harness, tool policy, reasoning effort, maximum output and treatment of failures. Some published tasks change installation checks or use a fallback score when a run fails. Those decisions are part of the measurement. Read the footnote for the specific benchmark before comparing this card with a report from another author.

Sources: https://huggingface.co/zai-org/GLM-5.3/blob/ba32f8c19c8738e2c57c71798c601d477a7bf844/README.md#L58

## Attributed configuration facts

```json
[
  {
    "name": "Reasoning effort",
    "value": "low / high / max; default max",
    "classification": "publisher_declaration",
    "evidence_url": "https://huggingface.co/zai-org/GLM-5.3/blob/ba32f8c19c8738e2c57c71798c601d477a7bf844/README.md#L55",
    "source_document_sha256": "ed1c0a4563c437a32f8953d637a1db9bd831de1b24bd3062a5ace9e04340b1cf"
  },
  {
    "name": "HLE-with-tools context / output cap",
    "value": "300000 / 163840 tokens",
    "classification": "publisher_declaration",
    "evidence_url": "https://huggingface.co/zai-org/GLM-5.3/blob/ba32f8c19c8738e2c57c71798c601d477a7bf844/README.md#L60",
    "source_document_sha256": "ed1c0a4563c437a32f8953d637a1db9bd831de1b24bd3062a5ace9e04340b1cf"
  },
  {
    "name": "DeepSWE context / timeout",
    "value": "400K tokens / 6 hours",
    "classification": "publisher_declaration",
    "evidence_url": "https://huggingface.co/zai-org/GLM-5.3/blob/ba32f8c19c8738e2c57c71798c601d477a7bf844/README.md#L62",
    "source_document_sha256": "ed1c0a4563c437a32f8953d637a1db9bd831de1b24bd3062a5ace9e04340b1cf"
  },
  {
    "name": "Terminal-Bench 3.0 repeats / turns / timeout",
    "value": "3 rollouts / 600 agent turns / 10 hours",
    "classification": "publisher_declaration",
    "evidence_url": "https://huggingface.co/zai-org/GLM-5.3/blob/ba32f8c19c8738e2c57c71798c601d477a7bf844/README.md#L64",
    "source_document_sha256": "ed1c0a4563c437a32f8953d637a1db9bd831de1b24bd3062a5ace9e04340b1cf"
  }
]
```

## Selected source-reported figures

Each point retains its own configuration, measurement unit and source. No site reproduction or cross-source ranking.

```json
[
  {
    "label": "DeepSWE v1.1",
    "metric": "DeepSWE",
    "value": 66.9,
    "unit": "reported score",
    "statistic": "reported value",
    "uncertainty": null,
    "conditions": {
      "model": "zai-org/GLM-5.3",
      "model_revision": "ba32f8c19c8738e2c57c71798c601d477a7bf844",
      "hardware": null,
      "executed_runtime_revision": null,
      "input_tokens": null,
      "output_tokens": null,
      "repetitions": null,
      "measurement_scope": "publisher capability evaluation, not local speed",
      "benchmark": "DeepSWE v1.1",
      "harness": "mini-swe-agent",
      "temperature": 0.95,
      "top_p": 1.0,
      "context": "400K",
      "timeout_hours": 6,
      "reasoning_effort": "max"
    },
    "evidence_state": "source_reported",
    "site_reproduced": false,
    "evidence_url": "https://huggingface.co/zai-org/GLM-5.3/blob/ba32f8c19c8738e2c57c71798c601d477a7bf844/README.md#L26",
    "source_document_sha256": "ed1c0a4563c437a32f8953d637a1db9bd831de1b24bd3062a5ace9e04340b1cf"
  },
  {
    "label": "Terminal-Bench 2.1",
    "metric": "Terminal_Bench",
    "value": 88.2,
    "unit": "reported score",
    "statistic": "reported value",
    "uncertainty": null,
    "conditions": {
      "model": "zai-org/GLM-5.3",
      "model_revision": "ba32f8c19c8738e2c57c71798c601d477a7bf844",
      "hardware": null,
      "executed_runtime_revision": null,
      "input_tokens": null,
      "output_tokens": null,
      "repetitions": null,
      "measurement_scope": "publisher capability evaluation, not local speed",
      "benchmark": "Terminal-Bench 2.1",
      "harness": "Claude Code 2.1.207",
      "temperature": 1.0,
      "top_p": 1.0,
      "max_new_tokens": 65536,
      "timeout_hours": 6,
      "reasoning_effort": "max"
    },
    "evidence_state": "source_reported",
    "site_reproduced": false,
    "evidence_url": "https://huggingface.co/zai-org/GLM-5.3/blob/ba32f8c19c8738e2c57c71798c601d477a7bf844/README.md#L24",
    "source_document_sha256": "ed1c0a4563c437a32f8953d637a1db9bd831de1b24bd3062a5ace9e04340b1cf"
  }
]
```