GRM 3.2 Cliff 9B

At the default Q4_K_M, 105 of 139 phones have enough estimated usable memory; 89 also reach 3+ tokens/s. On Mac, 52 of 54 exact configurations pass using that quant or a smaller tracked fallback.

Params
9.4B
Family
grm
Released
2026-07
Tags
chat · coding · vision · reasoning

Downloads & memory by quant

The download is only the weights — add KV cache and runtime to get what your phone actually needs. “Memory-fit” only means it can load; “usable” also requires an estimated 3+ tokens/s. Each row uses that row's quant. ★ = recommended.

QuantDownloadMemory neededPhone reality
Q2_K4.1 GB~5.6 GB129 memory-fit · 97 usable
Q3_K_M4.9 GB~6.4 GB105 memory-fit · 91 usable
IQ4_XS5.2 GB~6.7 GB105 memory-fit · 90 usable
Q4_05.5 GB~7 GB105 memory-fit · 89 usable
Q4_K_S5.6 GB~7.1 GB105 memory-fit · 89 usable
Q4_K_M5.9 GB~7.5 GB105 memory-fit · 89 usable
Q5_K_M6.9 GB~8.5 GB102 memory-fit · 86 usable
Q6_K7.7 GB~9.3 GB59 memory-fit · 59 usable
Q8_09.5 GB~11.2 GB59 memory-fit · 59 usable

GRM 3.2 Cliff 9B at Q4_K_M on 139 phones

This table is fixed to the recommended Q4_K_M, on each phone's largest RAM option. Counts for smaller quants above will differ. All phones are shown, sorted by fit and estimated experience; tap one for the full report.

PhoneChip · RAMNeedsSpeedVerdict

~ = bandwidth-based estimate · ✓ = measured on real hardware

GRM 3.2 Cliff 9B on 54 Mac configurations

Each Mac uses the recommended Q4_K_M at 4K context when it fits, or the largest smaller tracked quant that fits. Each row links to the full report with a macOS-specific setup guide.

MacUnified memoryQuantWorking budgetNeedsEstimated speedVerdict

FAQ

How much RAM do I need to run GRM 3.2 Cliff 9B on a phone?

At Q4_K_M, GRM 3.2 Cliff 9B needs ~7.5GB of usable memory (weights + KV cache + runtime). In practice that means a 12GB+ Android phone or a 12GB+ iPhone — iOS lets a single app use less of its RAM than Android does.

How fast does GRM 3.2 Cliff 9B run on a flagship phone?

On the OPPO Find X8 Pro (Dimensity 9400) it runs at ~6.5 tokens/s at Q4_K_M, estimated from memory bandwidth. Anything above ~8 tokens/s feels smooth for chat.

Can an iPhone run GRM 3.2 Cliff 9B?

Yes — the iPhone Air runs it at ~5.9 tokens/s at Q4_K_M, using 7.5GB of its ~7.8GB usable memory.

What is the best quantization of GRM 3.2 Cliff 9B for mobile?

Q4_K_M (5.9GB download) is the size/quality sweet spot of the 9 quants available. Total memory needed is ~7.5GB once the KV cache and runtime are counted — the download size alone understates it.

How many phones can run GRM 3.2 Cliff 9B?

105 of the 139 phones we track have enough estimated usable memory at Q4_K_M; 89 also reach our usable-speed threshold of 3 tokens/s, and 0 run smoothly. Memory-fit alone does not mean a good experience.

Which Macs can run GRM 3.2 Cliff 9B?

52 of the 54 exact Mac configurations we track have enough estimated working memory. Each Mac uses the recommended Q4_K_M when it fits, or the largest smaller tracked quant that fits, and separates installed unified memory, the macOS reserve, and estimated speed.

What is GRM 3.2 Cliff 9B good for on a phone?

It's tagged for chat, coding, vision, reasoning. At 9.4B parameters, it prioritizes answer quality over speed — expect slower decoding.

Similar models

grug 9B (ProCreations)Fara 1.5 9BQwen 3.5 9BOrnith 1.0 9BQwythos 9B v2