Run Ling-3.0 Flash on M4 Max

Ling-3.0 Flash runs at 73 tok/s on a 128 GB M4 Max, a community run reported by omlx.ai (MLX, 4-bit, 1k context). It needs about 96 GB of unified memory. It is Ant Group's 124B mixture-of-experts model (5B active), released under the MIT licence.

Speed
community73
tok/s
First Token
community1.145
seconds
RAM Needed
96
GB minimum
Engine
MLX
of this figure
Want to run Ling-3.0 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 secondcommunity73 tok/s
Provenancecommunity run · omlx.ai
Quantization4-bit
EngineMLX
Time to first token1.145s
Prompt processing (prefill)community1025 tok/s at 4K context
Context1k
Figure dated2026-11-06
Minimum RAM96 GB
Parameters124B MoE (5B active)
Chip bandwidth546 GB/s

Every figure for Ling-3.0 Flash on the M4 Max

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 (96 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
community73 tok/s4-bitMLX128 GB1komlx.ai40-core GPU. oMLX v0.7.2, MTP/DFlash off; Ling-3.0-flash-MLX-4bit; 66.8 GB peak. 72.4 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: Ling-3.0 Flash on the M4 Max

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 · 4-bit · 128 GB Mac · omlx.aicommunity73 tok/s
1k
community72.4 tok/scommunity1025 tok/s
at 4K
community1.145 s

Setup: run Ling-3.0 Flash on a Mac

Ollama — the index has not verified an Ollama tag for Ling-3.0 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:

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

Downloads: Search the Ollama library → · Hugging Face model card →

The 73 tok/s figure above came from MLX at 4-bit; 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.

Ling-3.0 Flash on other Apple Silicon chips

ChipSpeedQuantEngineBandwidth
M5 Maxcommunity68 tok/s4-bitMLX614 GB/s
M2 Maxcommunity44 tok/soQ4eMLX400 GB/s

About Ling-3.0 Flash

Ant Group's MIT-licensed 124B MoE with 5B active and 256K context. About 67 GB in use at 4-bit, so it needs a 96 GB+ Mac; a community oMLX run measured 68 tok/s on an M5 Max (MLX 4-bit, MTP off).

MIT licence · 124B MoE · 256K context · LLMCheck capability 38/50 · see its leaderboard rank →

Memory by context length

Estimated, not measured: the index’s minimum for Ling-3.0 Flash (96 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 7 of 42 layers (every 6th, per modeling_bailing_moe_v3.py); the rest are Kimi Delta Attention (linear, constant state). 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 M4 Max Mac on macOS 15, the lower default; macOS 26 allows more): “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~0.1 GB~96.0 GB~96.0 GB128 GB Mac raise GPU limit (macOS 15)192 GB Mac
32K~0.3 GB~96.2 GB~96.1 GB128 GB Mac raise GPU limit (macOS 15)192 GB Mac
64K (coding-agent context)~0.5 GB~96.5 GB~96.2 GB128 GB Mac raise GPU limit (macOS 15)192 GB Mac
128K~1.1 GB~97.0 GB~96.5 GB128 GB Mac raise GPU limit (macOS 15)192 GB Mac

System requirements

Macs with the M4 Max

🛒 Get a Mac that runs Ling-3.0 Flash

Ling-3.0 Flash needs about 96 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, so the used option links Amazon Renewed listings.

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 Ling-3.0 Flash stacks up against other models on your specific Mac hardware.

Open Compare Tool Full Leaderboard