Hardware / Apple /iPhone 15 Pro

Apple · Phone

iPhone 15 Pro

Runs 65 models fully in memory, 0 more with CPU offload. 3 of those sit close enough to 4.4 GB that our size estimate is what decides them.

Memory one app may use

4.4 GB of 8 GB

the most one app may use, not the memory in the phone — estimated

Bandwidth

60 GB/s

Runs

65

01

Specifications

Memory one app may use

4.4 GB of 8 GB

What one app is allowed, not the whole device.

Memory bandwidth

60 GB/s

How fast it can read the model. Sets the speed of the reply.

Released

Sep 22, 2023

Architecture

A17 Pro

02

What it runs

Best quant per model. Text models first, then speech, then what does not fit.

What is quantisation? →

ModelParamsContextSpeedVerdict

Llama 3.2 1B Instruct1.2B33K37 tok/sFits in memory Llama 3.2 3B Instruct3.2B33K14 tok/sFits in memory Hy-MT2-1.8B2B8K19 tok/sFits in memory Granite 4.0 Micro3.2B8K14 tok/sFits in memory Gemma 3 4B4.3B8K10 tok/sFits in memoryest Ministral 3 3B 25123.8B4K12 tok/sFits in memory Nemotron 3.5 Content Safety4.3B2K10 tok/sFits in memoryest

Speech models it runs58 modelsTranscription and speech models that fit in memory. Throughput figures for a transcription model come from its text decoder, so treat this as a fit answer rather than a speed one.

Niagara 19m Batch.en20M2227 tok/sFits in memory Niagara 38m Batch.en38M1172 tok/sFits in memory Hubert Large Ls960 ft0.3B119 tok/sFits in memory Wav2vec2 Large 960h Lv60 Self0.3B139 tok/sFits in memory Canary 180m Flash0.2B211 tok/sFits in memory Parakeet TDT CTC 110m0.1B405 tok/sFits in memory Zipformer cr CTC Transducer XL 290M0.3B131 tok/sFits in memory Zipformer Transducer XL 290M0.3B154 tok/sFits in memory Niagara 9m Batch.en9M4218 tok/sFits in memory Niagara 84m Batch.en84M303 tok/sFits in memory Audio8 ASR 0.1B0.3B33K149 tok/sFits in memory ASR Conformer Largescaleasr0.5B93 tok/sFits in memory Granite Speech 5.0 470m Turboctc0.5B89 tok/sFits in memory Granite Speech 5.0 470m Turboctc nc0.5B89 tok/sFits in memory Nemotron 3.5 ASR Streaming 0.6b0.6B74 tok/sFits in memory Nemotron Speech Streaming en 0.6b0.6B74 tok/sFits in memory Parakeet CTC 0.6b0.6B74 tok/sFits in memory Parakeet RNNT 0.6b0.6B74 tok/sFits in memory Parakeet TDT 0.6b v20.6B74 tok/sFits in memory Parakeet TDT 0.6b v30.6B74 tok/sFits in memory Moonshine Streaming Medium0.3B4K127 tok/sFits in memory Moonshine Streaming Small0.1B4K380 tok/sFits in memory Distil Large v3.50.8B56 tok/sFits in memory Canary 1b Flash0.8B56 tok/sFits in memory Whisper Large v3 Turbo0.8B56 tok/sFits in memory Qwen3 ASR 0.6B0.9B50 tok/sFits in memory Owsm CTC v3.2 ft 1B1B45 tok/sFits in memory Owsm CTC v4 1B1B45 tok/sFits in memory MMS 1b All1B45 tok/sFits in memory Canary 1b1B45 tok/sFits in memory Canary 1b v21B45 tok/sFits in memory Owsm CTC v3.1 1B1.1B41 tok/sFits in memory Parakeet CTC 1.1b1.1B41 tok/sFits in memory Parakeet RNNT 1.1b1.1B41 tok/sFits in memory Parakeet TDT 1.1b1.1B41 tok/sFits in memory Lite Whisper Large v3 Acc1.4B32 tok/sFits in memory Qwen3 ASR 0.6B HF0.8B56 tok/sFits in memory Fun ASR Nano 2512 HF0.8B56 tok/sFits in memory Nyra Health CrisperWhisper1.6B28 tok/sFits in memory Whisper Large v31.5B30 tok/sFits in memory ARK ASR 0.6B1.3B33K34 tok/sFits in memory MOSS Transcribe Diarize0.9B50 tok/sFits in memory Cohere Transcribe 03 20262.1B21 tok/sFits in memory Qwen3 ASR 1.7B2.3B19 tok/sFits in memory STT 2.6b en2.6B15 tok/sFits in memory Canary Qwen 2.5b2.6B17 tok/sFits in memory Qwen3 ASR 1.7B HF2B22 tok/sFits in memory Granite 4.0 1b Speech2.3B19 tok/sFits in memory Granite Speech 4.1 2b2.3B19 tok/sFits in memory Granite Speech 4.1 2b NAR2.3B19 tok/sFits in memory GLM ASR Nano 25122.3B19 tok/sFits in memory Higgs Audio v3 STT2.7B17 tok/sFits in memory Granite Speech 3.3 2b3B15 tok/sFits in memory ARK ASR 3B4.1B8K11 tok/sFits in memory MOSS Transcribe Preview 2B2.4B2K19 tok/sFits in memory Voxtral Mini 4B Realtime 26024.4B10 tok/sFits in memoryest Moonshine Streaming Tiny0B4Kno speed estimateFits in memory Moonshine Tiny0B194no speed estimateFits in memory

Too large for this device159 modelsThe weights do not fit even with part of them offloaded to system memory. Closest calls first.

Gemma 4 E2B it5.1B—no speed estimateToo largeest Hojo ASR V15.2B—no speed estimateToo large Phi 4 Multimodal Instruct5.6B—no speed estimateToo large Reka Edge7.1B—no speed estimateToo large Llama 3.1 8B Instruct8B—no speed estimateToo large Llama 3 8B Lunaris8B—no speed estimateToo large VibeVoice ASR HF8.3B—no speed estimateToo large Voxtral Mini 3B 25074.7B—no speed estimateToo large Qwen3 8B8.2B—no speed estimateToo large Qwen3 VL 8B Thinking8.8B—no speed estimateToo large Gemma 3n 4B7.8B—no speed estimateToo large Granite 4.2 8B8.8B—no speed estimateToo large Hy-MT2-7B8B—no speed estimateToo large Qwen3 VL 8B Instruct8.8B—no speed estimateToo large Granite 4.1 8B8.8B—no speed estimateToo large Ministral 3 8B 25128.9B—no speed estimateToo large Gemma 3 12B12.2B—no speed estimateToo large Rocinante 12B12.2B—no speed estimateToo large Qwen2.5 7B Instruct7.6B—no speed estimateToo large Llama Guard 4 12B12B—no speed estimateToo large Gemma 4 E4B it8B—no speed estimateToo large Ministral 3 14B 251213.9B—no speed estimateToo large UI-TARS 7B8.3B—no speed estimateToo large UnslopNemo 12B12.2B—no speed estimateToo large Gemma 4 12B it12B—no speed estimateToo large Granite Speech 3.3 8b8.6B—no speed estimateToo large Higgs Audio v3 8b STT v28.9B—no speed estimateToo large Qwen3.5-9B9.7B—no speed estimateToo large ReMM SLERP 13B13B—no speed estimateToo large Qwen3 14B14.8B—no speed estimateToo large Mistral Nemo12.2B—no speed estimateToo large Cydonia 24B V4.123.6B—no speed estimateToo large Mistral Small 323.6B—no speed estimateToo large Uncensored24B—no speed estimateToo large Mistral Small 3.1 24B24B—no speed estimateToo large Voxtral Small 24B 250724.3B—no speed estimateToo large MythoMax 13B13B—no speed estimateToo large Ternary Bonsai 2 27B27B—no speed estimateToo large Muse Glimmer 30B29.8B—no speed estimateToo large Qwen3 30B A3B Instruct 250730.5B—no speed estimateToo large Gemma 4 26B A4B25.8B—no speed estimateToo large Qwen3 VL 30B A3B Instruct31.1B—no speed estimateToo large Nemotron 3 Nano 30B A3B31.6B—no speed estimateToo large Gemma 3 27B27.4B—no speed estimateToo large Qwen3.8 27B27.8B—no speed estimateToo large Gemma 4 31B31.3B—no speed estimateToo large Nex-N2.5-Mini35.1B—no speed estimateToo large Qwen3.5-35B-A3B36B—no speed estimateToo large Qwen3 VL 30B A3B Thinking31.1B—no speed estimateToo large Reka Flash 320.9B—no speed estimateToo large Nemotron 3.5 Lightning31.6B—no speed estimateToo large gpt-oss-safeguard-20b21.5B—no speed estimateToo large Qwen3 32B32.8B—no speed estimateToo large Qwen3.6 35B A3B36B—no speed estimateToo large Mistral Small 3.2 24B24B—no speed estimateToo large Skyfall 36B V236.9B—no speed estimateToo large Phi 414.7B—no speed estimateToo large Gemma 2 27B27.2B—no speed estimateToo large Qwen3.5-27B27.8B—no speed estimateToo large Qwen3.6 27B27.8B—no speed estimateToo large GLM 4.7 Flash31.2B—no speed estimateToo large Hy-MT2-30B-A3B30.1B—no speed estimateToo large Qwen3 30B A3B Thinking 250730.5B—no speed estimateToo large Qwen3 Coder 30B A3B Instruct30.5B—no speed estimateToo large Laguna XS 2.133.4B—no speed estimateToo large Olmo 3 32B Think32.2B—no speed estimateToo large Qwen2.5 Coder 32B Instruct32.8B—no speed estimateToo large Qwen3 VL 32B Instruct33.4B—no speed estimateToo large Nex-N2-Mini35.1B—no speed estimateToo large Llama 3.1 70B Instruct70.6B—no speed estimateToo large R1 Distill Llama 70B70.6B—no speed estimateToo large Qwen3 Next 80B A3B Instruct81.3B—no speed estimateToo large Hermes 3 70B Instruct70.6B—no speed estimateToo large Qwen2.5 72B Instruct72.7B—no speed estimateToo large Hunyuan A13B Instruct80.4B—no speed estimateToo large Qwen3 Next 80B A3B Thinking81.3B—no speed estimateToo large Command A111B—no speed estimateToo large gpt-oss-120b120B—no speed estimateToo large Mistral Small 4119B—no speed estimateToo large Nemotron 3 Super124B—no speed estimateToo large Llama 3.3 70B Instruct70.6B—no speed estimateToo large Hermes 4 70B70.6B—no speed estimateToo large Llama 3.3 Euryale 70B70.6B—no speed estimateToo large Llama 3.1 Euryale 70B v2.270.6B—no speed estimateToo large GLM 4.6V108B—no speed estimateToo large GLM 4.5V108B—no speed estimateToo large Ling-2.6-flash107B—no speed estimateToo large Llama 4 Scout109B—no speed estimateToo large Magnum v4 72B72.7B—no speed estimateToo large Qwen2.5 VL 72B Instruct73.4B—no speed estimateToo large Qwen3 Coder Next79.7B—no speed estimateToo large Laguna S 2.1118B—no speed estimateToo large Ling-3.0-flash128B—no speed estimateToo large Mixtral 8x22B Instruct141B—no speed estimateToo large GLM 4.5 Air111B—no speed estimateToo large Step 3.7 Flash201B—no speed estimateToo large Qwen3.5-122B-A10B125B—no speed estimateToo large Devstral 2 2512125B—no speed estimateToo large Mistral Medium 3.5128B—no speed estimateToo large MiniMax M2229B—no speed estimateToo large Step 3.5 Flash199B—no speed estimateToo large MiniMax M2.5229B—no speed estimateToo large Qwen3 235B A22B Thinking 2507235B—no speed estimateToo large Qwen3 VL 235B A22B Thinking236B—no speed estimateToo large Qwen3 VL 235B A22B Instruct236B—no speed estimateToo large DeepSeek V4 Flash291B—no speed estimateToo large Hy3299B—no speed estimateToo large GLM 4.5358B—no speed estimateToo large Laguna M.1226B—no speed estimateToo large Trinity Large Thinking399B—no speed estimateToo large MiniMax M2.7229B—no speed estimateToo large MiniMax M2.1229B—no speed estimateToo large Llama 4 Maverick402B—no speed estimateToo large Qwen3 235B A22B Instruct 2507235B—no speed estimateToo large GLM 4.6357B—no speed estimateToo large GLM 4.7358B—no speed estimateToo large MiniMax M3427B—no speed estimateToo large MiniMax-01456B—no speed estimateToo large Nex-N2.5-Pro397B—no speed estimateToo large Inkling Small266B—no speed estimateToo large Jamba Large 1.7399B—no speed estimateToo large ERNIE 4.5 VL 424B A47B424B—no speed estimateToo large Hy3 preview299B—no speed estimateToo large DeepSeek V4 Flash Vision Exp305B—no speed estimateToo large MiniMax M1456B—no speed estimateToo large MiMo-V2.5311B—no speed estimateToo large GLM 5.3 Flash321B—no speed estimateToo large Qwen3 Coder 480B A35B480B—no speed estimateToo large DeepSeek V3.2685B—no speed estimateToo large Nex-N2-Pro397B—no speed estimateToo large Qwen3.5 397B A17B403B—no speed estimateToo large Hermes 4 405B406B—no speed estimateToo large Hermes 3 405B Instruct406B—no speed estimateToo large GLM 5.1754B—no speed estimateToo large DeepSeek V4.1 Flash763B—no speed estimateToo large R1 0528685B—no speed estimateToo large DeepSeek V3685B—no speed estimateToo large DeepSeek V3.1685B—no speed estimateToo large GLM 5754B—no speed estimateToo large GLM 5.2753B—no speed estimateToo large Nemotron 3 Ultra561B—no speed estimateToo large MiMo-V2.5-Pro1T—no speed estimateToo large Kimi K2 09051T—no speed estimateToo large Kimi K2 07111T—no speed estimateToo large Ling-2.6-1T1T—no speed estimateToo large DeepSeek V3.1 Terminus685B—no speed estimateToo large DeepSeek V3 0324685B—no speed estimateToo large DeepSeek V3.2 Exp685B—no speed estimateToo large Ring-2.6-1T1T—no speed estimateToo large Kimi K2.7 Code1.1T—no speed estimateToo large Kimi K2.61.1T—no speed estimateToo large Kimi K2.51.1T—no speed estimateToo large Hy4 preview780B—no speed estimateToo large DeepSeek V4 Pro1.6T—no speed estimateToo large Inkling952B—no speed estimateToo large Kimi K2 Thinking1.1T—no speed estimateToo large LongCat 2.01.8T—no speed estimateToo large Qwen3.8 2.4T A95B2.4T—no speed estimateToo large Kimi K32.8T—no speed estimateToo large

224 models have a verdict on this device: 65 fit in memory, 0 spill into system memory and 159 are too large. Every one of them is on this page.

Something wrong on this page? Tell us