The LLMCheck index estimates GLM-5.3-Flash at ~27 tok/s on an M5 Max at IQ3_XXS, a figure from its published bandwidth model, not a measured run. It needs about 128 GB of unified memory. It is Zhipu AI's 320B mixture-of-experts model (18B active), released under the MIT licence.
Want to run GLM-5.3-Flash faster — or step up to bigger models? Find your Mac →This is an LLMCheck index estimate from its published two-term bandwidth model, not a measured run; the index labels it so everywhere it appears. Own this Mac? Submit a real run →
| Metric | Value |
|---|---|
| Tokens per second | estimated~27 tok/s |
| Provenance | estimate · how estimates are made |
| Quantization | IQ3_XXS |
| Engine | MLX |
| Figure dated | 2026-09 |
| Minimum RAM | 128 GB |
| Parameters | 320B MoE (18B active) |
| Chip bandwidth | 614 GB/s |
Each row as the index holds it, with its source. “Mac RAM” is the memory of the Mac a run used, not the model’s requirement (128 GB). The headline figure follows the reference rule: the lower-median measured plain-decode run at 5 bits or fewer (a run whose quant was not stated counts only when no stated-quant run qualifies), else the estimate; a run marked “not used to anchor estimates” is listed here but never becomes the headline.
| Speed | Quant | Engine | Mac RAM | Context | Source | Note |
|---|---|---|---|---|---|---|
| estimated27 tok/s | IQ3_XXS | MLX | 128 GB | — | LLMCheck estimator | — |
Ollama — the index has not verified an Ollama tag for GLM-5.3-Flash, so it doesn’t print one (a guessed tag fails with “pull model manifest: file does not exist”). Search the Ollama library →
MLX — pick a build from Hugging Face’s MLX listings, then run it with its repo id:
Downloads: Search the Ollama library → · model card →
Prefer a GUI? LM Studio downloads GGUF and MLX builds in one click. Step by step: install Ollama on a Mac.
Zhipu's 320B MoE with 18B active — the model that ran anonymously as 'Ox Alpha' until 2026-08-26. MIT, natively multimodal, hybrid sparse + linear attention, 300K context. Terminal-Bench 2.1 84.3 (model card). 3-bit is 120 GB and just fits a 128 GB Mac; 4-bit needs 192 GB+. ~27 tok/s estimated on M5 Max at 3-bit, ~35 on M5 Ultra, using the index's MoE factor. Supported in upstream llama.cpp since 30 Sep 2026 (glm5-next, with Metal), and through MLX community builds.
MIT licence · 320B MoE · 300K context · LLMCheck capability 48/50 · see its leaderboard rank →
GLM-5.3-Flash needs about 128 GB of unified memory. These current Apple Silicon Macs have the headroom to run it comfortably:
As an Amazon Associate, LLMCheck earns from qualifying purchases. These affiliate links cost you nothing extra and help keep the LLMCheck index free.
See how GLM-5.3-Flash stacks up against other models on your specific Mac hardware.
Open Compare Tool Full Leaderboard