Run GLM-4.7-Flash on M6

GLM-4.7-Flash runs at 47 tok/s (46.6 as recorded) on a 32 GB M6, a community run reported by omlx.ai (MLX, oQ4e, 1k context). It needs about 18 GB of unified memory. It is Zhipu AI's 31B mixture-of-experts model (3B active), released under the MIT licence.

Speed
community47
tok/s
First Token
community1.092
seconds
RAM Needed
18
GB minimum
Engine
MLX
of this figure
Want to run GLM-4.7-Flash faster — or step up to bigger models? Find your Mac →

The figure

This is a community run: a real run, reported by its source and linked below. The index records it as given and never adjusts it.

MetricValue
Tokens per secondcommunity47 tok/s
Provenancecommunity run · omlx.ai
QuantizationoQ4e
EngineMLX
Time to first token1.092s
Prompt processing (prefill)community893 tok/s at 4K context
Context1k
Figure dated2026-11-11
Minimum RAM18 GB
Parameters31B MoE (3B active)
Chip bandwidth170 GB/s

Every figure for GLM-4.7-Flash on the M6

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 (18 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.

SpeedQuantEngineMac RAMContextSourceNote
community46.6 tok/soQ4eMLX32 GB1komlx.ai12-core GPU. oMLX v0.7.4.dev1, MTP/DFlash off; GLM-4.7-Flash-oQ4e-fp16-mtp (MTP head not used); 17.0 GB peak. 37.6 tok/s at 4k.
Provenance key
sourced
Published figure — vendor or press review
estimated
community
Linked individual run — blog, forum or submission

Speed by context depth: GLM-4.7-Flash on the M6

Measured figures only, each from its source’s own session; the index does not estimate long-context or prefill speed, so a missing cell is —. Each @ column is plain decode with that many tokens already in the prompt, from the same session as the row’s short-context figure: decode slows as the KV cache it reads grows. Prefill is prompt processing, the tokens per second a runtime reads a prompt at; with long agent contexts it sets how long you wait before the first token.

RunDecode (row)@4KPrefillTTFT
MLX · oQ4e · 32 GB Mac · omlx.aicommunity46.6 tok/s
1k
community37.6 tok/scommunity893 tok/s
at 4K
community1.092 s

Setup: run GLM-4.7-Flash on a Mac

Ollama — install it (brew install ollama or the app from ollama.com), then pull the model by its tag, checked on ollama.com on 2026-11-14:

ollama run glm-4.7-flash

MLX — pick a build from Hugging Face’s MLX listings, then run it with its repo id:

pip install -U mlx-lm
mlx_lm.generate --model <repo-id> --prompt "Hello"

Downloads: Ollama page → · Hugging Face model card →

The 47 tok/s figure above came from MLX at oQ4e; other runtimes and quants run at different speeds.

Prefer a GUI? LM Studio downloads GGUF and MLX builds in one click. Step by step: install Ollama on a Mac.

GLM-4.7-Flash on other Apple Silicon chips

ChipSpeedQuantEngineBandwidth
M5 Maxestimated107 tok/sQ4_K_MMLX614 GB/s
M4 Maxestimated102 tok/sQ4_K_MMLX546 GB/s
M4 Proestimated68 tok/sQ4_K_MOllama273 GB/s

For the estimated figures here, the bandwidth formula’s typical error for small MoEs (≤40B total) is ±15%, against 88 measured runs of 15 models (how it is computed).

About GLM-4.7-Flash

Zhipu AI's MIT-licensed 31B MoE with ~3B active (19 Jan 2026; its card calls it a 30B-A3B model), 200K context. Until November 2026 the index listed it as a dense 31B model, which understated its speed several-fold. A community oMLX run measured 46.6 tok/s on a 32 GB Mac mini M6 (oQ4e, MTP off).

MIT licence · 31B MoE · 200K context · LLMCheck capability 35/50 · see its leaderboard rank →

Memory by context length

Estimated, not measured: the index’s minimum for GLM-4.7-Flash (18 GB — the weights at the quant that fits, plus ~2 GB) plus the KV cache for that many tokens, computed from its config.json. MLA on all 47 layers (kv_lora_rank 512 + rope 64); runtimes that expand MLA to per-head K/V use far more. A 16-bit KV cache is the default in llama.cpp, Ollama, LM Studio and mlx-lm; a q8 cache halves the KV part. Coding agents hold 64K tokens or more. Runtimes also add a buffer that grows with context, so leave a few GB spare. macOS lets the GPU use only part of a Mac’s memory by default (figures for an M6 Mac, which runs macOS 26 or later): “fits” means within the GPU memory a Mac of that tier can use with the limit raised, a tier tagged raise GPU limit fits only after sudo sysctl iogpu.wired_limit_mb raises it (the tooltip has the exact command), and the last column is the smallest Mac memory size whose default limit holds it (the GPU memory limit).

ContextKV cache (16-bit)RAM neededRAM, q8 KVSmallest tier that fitsWithin the default GPU limit
8K~0.4 GB~18.2 GB~18.1 GB24 GB Mac raise GPU limit32 GB Mac
32K~1.8 GB~19.6 GB~18.8 GB24 GB Mac raise GPU limit32 GB Mac
64K (coding-agent context)~3.5 GB~21.3 GB~19.7 GB32 GB Macsame
128K~7.1 GB~24.9 GB~21.4 GB32 GB Macsame

System requirements

Macs with the M6

🛒 Get a Mac that runs GLM-4.7-Flash

GLM-4.7-Flash needs about 18 GB of unified memory. These Apple Silicon Macs have enough to run it. Apple builds the higher-memory configurations to order; Amazon does not stock them.

As an Amazon Associate, LLMCheck earns from qualifying purchases. These affiliate links cost you nothing extra and help keep the LLMCheck index free.

Compare More Models

See how GLM-4.7-Flash stacks up against other models on your specific Mac hardware.

Open Compare Tool Full Leaderboard