Run Raon-1 mini preview on M2 Max

Raon-1 mini preview runs at 68 tok/s (68.4 as recorded) on a 96 GB M2 Max, a community run reported by omlx.ai (MLX, NVFP4, 1k context). It needs about 18 GB of unified memory. It is KRAFTON's 28.8B mixture-of-experts model (4.3B active), released under the Apache 2.0 licence.

Speed
community68
tok/s
First Token
community1.062
seconds
RAM Needed
18
GB minimum
Engine
MLX
of this figure
Want to run Raon-1 mini preview 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 secondcommunity68 tok/s
Provenancecommunity run · omlx.ai
QuantizationNVFP4
EngineMLX
Time to first token1.062s
Prompt processing (prefill)community947 tok/s at 4K context
Context1k
Figure dated2026-11-14
Minimum RAM18 GB
Parameters28.8B MoE (4.3B active)
Chip bandwidth400 GB/s

Every figure for Raon-1 mini preview on the M2 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 (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
community68.4 tok/sNVFP4MLX96 GB1komlx.ai38-core GPU. oMLX v0.8.0rc1 with float16 activations, MTP/DFlash/SpecPrefill off; KRAFTON Raon-1-mini-preview-mlx (NVFP4 experts, MXFP8 output head; the page lists its quant as unknown); 18.5 GB peak. 62.2 tok/s at 4k. An earlier session on 9 Nov (v0.7.4.dev1, tr6q86gs) read 66.8.
Provenance key
sourced
Published figure — vendor or press review
estimated
community
Linked individual run — blog, forum or submission

Speed by context depth: Raon-1 mini preview on the M2 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 · NVFP4 · 96 GB Mac · omlx.aicommunity68.4 tok/s
1k
community62.2 tok/scommunity947 tok/s
at 4K
community1.062 s

Setup: run Raon-1 mini preview on a Mac

Ollama — the index has not verified an Ollama tag for Raon-1 mini preview, 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: Hugging Face repo → · Hugging Face model card →

The 68 tok/s figure above came from MLX at NVFP4; 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.

Raon-1 mini preview on other Apple Silicon chips

ChipSpeedQuantEngineBandwidth
M5 Maxestimated123 tok/sQ4_K_Mllama.cpp614 GB/s
M4 Maxcommunity114 tok/soQ4eMLX546 GB/s
M5 Procommunity86 tok/sNVFP4MLX307 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 Raon-1 mini preview

KRAFTON's Apache 2.0 preview MoE trained from scratch (21 Oct 2026): 28.8B total, 4.3B active, 262K context, text only, Korean-first across seven languages, thinking mode by default. KRAFTON's card (preview, vendor-run, GPT judge) reports GPQA Diamond 84.3, LiveCodeBench v6 81.6 and SWE-bench Verified 70.8 on mini-SWE-agent, with MMLU-Pro 76.5 below its Qwen 3.6 and Gemma 4 comparators. Official GGUFs start at 16.1 GB (IQ4_XS), so it targets 24 GB Macs. It runs in oMLX 0.7.3+ and mlx-vlm 0.7.12+ without remote code (upstream llama.cpp support is KRAFTON's open PR #32176), and community oMLX runs measured 114–129 tok/s on an M4 Max (MTP off); specs may change before the final release.

Apache 2.0 licence · 28.8B MoE · 262K context · LLMCheck capability 34/50 · see its leaderboard rank →

Memory by context length

Estimated, not measured: the index’s minimum for Raon-1 mini preview (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 25 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 M2 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~0.2 GB~18.1 GB~18.1 GB24 GB Mac raise GPU limit32 GB Mac
32K~0.9 GB~18.8 GB~18.4 GB24 GB Mac raise GPU limit32 GB Mac
64K (coding-agent context)~1.9 GB~19.8 GB~18.9 GB24 GB Mac raise GPU limit32 GB Mac
128K~3.8 GB~21.7 GB~19.8 GB32 GB Mac raise GPU limit (macOS 15)36 GB Mac

System requirements

Macs with the M2 Max

🛒 Get a Mac that runs Raon-1 mini preview

Raon-1 mini preview 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 Raon-1 mini preview stacks up against other models on your specific Mac hardware.

Open Compare Tool Full Leaderboard