Hardware / NVIDIA /GeForce RTX 4070 Ti SUPER

NVIDIA · GPU

GeForce RTX 4070 Ti SUPER

Runs 98 models fully in memory, 25 more with CPU offload.

Memory

16 GB

Bandwidth

672 GB/s

Runs

98

01

Specifications

Memory

16 GB

Decides which models fit at all.

Memory bandwidth

672 GB/s

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

Compute

44.1 TFLOPS

Trillions of half-precision sums a second. Sets how fast it reads a long prompt, not how fast it replies.

Power draw

285 W

At full load, for sizing a power supply.

MSRP

$799

What the maker asked for it new. We hold no used or street prices, so this is not what one costs today.

Released

Jan 24, 2024

Architecture

Ada Lovelace

02

What it runs

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

What is quantisation? →

ModelParamsContextSpeedVerdict

Ministral 3 3B 25123.8B66K233 tok/sFits in memory Granite 4.0 Micro3.2B66K158 tok/sFits in memory Llama 3.2 3B Instruct3.2B131K236 tok/sFits in memory Gemma 3 4B4.3B131K176 tok/sFits in memory Nemotron 3.5 Content Safety4.3B66K176 tok/sFits in memory Llama 3.1 8B Instruct8B131K111 tok/sFits in memory UI-TARS 7B8.3B66K107 tok/sFits in memory Llama 3.2 1B Instruct1.2B33K739 tok/sFits in memory Qwen2.5 7B Instruct7.6B33K117 tok/sFits in memory Gemma 3n 4B7.8B33K114 tok/sFits in memory Reka Edge7.1B16K125 tok/sFits in memory Qwen3 8B8.2B33K108 tok/sFits in memory Granite 4.1 8B8.8B33K101 tok/sFits in memory Ministral 3 8B 25128.9B33K100 tok/sFits in memory Qwen3 VL 8B Thinking8.8B33K101 tok/sFits in memory Qwen3 VL 8B Instruct8.8B33K101 tok/sFits in memory Granite 4.2 8B8.8B33K101 tok/sFits in memory Hy-MT2-1.8B2B8K215 tok/sFits in memory Qwen3.5-9B9.7B33K91 tok/sFits in memory Llama 3 8B Lunaris8B8K111 tok/sFits in memory Llama Guard 4 12B12B66K74 tok/sFits in memory Gemma 3 12B12.2B66K73 tok/sFits in memory UnslopNemo 12B12.2B33K73 tok/sFits in memory Rocinante 12B12.2B33K73 tok/sFits in memory Mistral Nemo12.2B33K73 tok/sFits in memory Hy-MT2-7B8B8K94 tok/sFits in memory Ministral 3 14B 251213.9B16K64 tok/sFits in memory Qwen3 14B14.8B16K60 tok/sFits in memory Phi 414.7B16K60 tok/sFits in memory ReMM SLERP 13B13B4K68 tok/sFits in memory MythoMax 13B13B4K68 tok/sFits in memory

Speech models it runs67 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.en20M25299 tok/sFits in memory Niagara 38m Batch.en38M13315 tok/sFits in memory Higgs Audio v3 STT2.7B187 tok/sFits in memory Cohere Transcribe 03 20262.1B241 tok/sFits in memory Distil Large v3.50.8B633 tok/sFits in memory Lite Whisper Large v3 Acc1.4B540 tok/sFits in memory Owsm CTC v3.1 1B1.1B806 tok/sFits in memory Owsm CTC v3.2 ft 1B1B506 tok/sFits in memory Owsm CTC v4 1B1B887 tok/sFits in memory Hubert Large Ls960 ft0.3B1581 tok/sFits in memory MMS 1b All1B756 tok/sFits in memory Wav2vec2 Large 960h Lv60 Self0.3B1581 tok/sFits in memory Hojo ASR V15.2B146 tok/sFits in memory Granite 4.0 1b Speech2.3B220 tok/sFits in memory Granite Speech 3.3 2b3B296 tok/sFits in memory Granite Speech 4.1 2b2.3B386 tok/sFits in memory Granite Speech 4.1 2b NAR2.3B386 tok/sFits in memory STT 2.6b en2.6B341 tok/sFits in memory Phi 4 Multimodal Instruct5.6B66K158 tok/sFits in memory Voxtral Mini 3B 25074.7B161 tok/sFits in memory Voxtral Mini 4B Realtime 26024.4B172 tok/sFits in memory Canary 180m Flash0.2B4199 tok/sFits in memory Canary 1b1B506 tok/sFits in memory Canary 1b Flash0.8B633 tok/sFits in memory Canary 1b v21B506 tok/sFits in memory Canary Qwen 2.5b2.6B195 tok/sFits in memory Nemotron 3.5 ASR Streaming 0.6b0.6B843 tok/sFits in memory Nemotron Speech Streaming en 0.6b0.6B843 tok/sFits in memory Parakeet CTC 0.6b0.6B1478 tok/sFits in memory Parakeet CTC 1.1b1.1B460 tok/sFits in memory Parakeet RNNT 0.6b0.6B843 tok/sFits in memory Parakeet RNNT 1.1b1.1B806 tok/sFits in memory Parakeet TDT CTC 110m0.1B4600 tok/sFits in memory Parakeet TDT 0.6b v20.6B1260 tok/sFits in memory Parakeet TDT 0.6b v30.6B843 tok/sFits in memory Parakeet TDT 1.1b1.1B460 tok/sFits in memory Nyra Health CrisperWhisper1.6B316 tok/sFits in memory Whisper Large v31.5B504 tok/sFits in memory Whisper Large v3 Turbo0.8B1109 tok/sFits in memory MOSS Transcribe Diarize0.9B562 tok/sFits in memory Qwen3 ASR 0.6B0.9B985 tok/sFits in memory Qwen3 ASR 0.6B HF0.8B1109 tok/sFits in memory Qwen3 ASR 1.7B2.3B329 tok/sFits in memory Qwen3 ASR 1.7B HF2B253 tok/sFits in memory Zipformer cr CTC Transducer XL 290M0.3B2606 tok/sFits in memory Zipformer Transducer XL 290M0.3B1745 tok/sFits in memory ASR Conformer Largescaleasr0.5B1847 tok/sFits in memory GLM ASR Nano 25122.3B329 tok/sFits in memory Granite Speech 5.0 470m Turboctc0.5B1774 tok/sFits in memory Granite Speech 5.0 470m Turboctc nc0.5B1512 tok/sFits in memory Fun ASR Nano 2512 HF0.8B1109 tok/sFits in memory Gemma 4 E2B it5.1B174 tok/sFits in memory Niagara 9m Batch.en9M56220 tok/sFits in memory Niagara 84m Batch.en84M10557 tok/sFits in memory Gemma 4 E4B it8B66K111 tok/sFits in memory ARK ASR 0.6B1.3B33K581 tok/sFits in memory ARK ASR 3B4.1B33K216 tok/sFits in memory Audio8 ASR 0.1B0.3B33K1687 tok/sFits in memory VibeVoice ASR HF8.3B107 tok/sFits in memory MOSS Transcribe Preview 2B2.4B16K370 tok/sFits in memory Granite Speech 3.3 8b8.6B103 tok/sFits in memory Higgs Audio v3 8b STT v28.9B100 tok/sFits in memory Moonshine Streaming Medium0.3B4K2956 tok/sFits in memory Moonshine Streaming Small0.1B4K5060 tok/sFits in memory Gemma 4 12B it12B74 tok/sFits in memory Moonshine Streaming Tiny0B4Kno speed estimateFits in memory Moonshine Tiny0B194no speed estimateFits in memory

Runs only by spilling into system memory25 modelsThese load, but part of the weights sits in ordinary system memory, which is far slower than the chip.

Gemma 4 26B A4B25.8B—no speed estimateSpills to system RAM Qwen3.5-27B27.8B—no speed estimateSpills to system RAM GLM 4.7 Flash31.2B—no speed estimateSpills to system RAM Nemotron 3 Nano 30B A3B31.6B—no speed estimateSpills to system RAMest Voxtral Small 24B 250724.3B—no speed estimateSpills to system RAM gpt-oss-safeguard-20b21.5B—no speed estimateSpills to system RAM Qwen3 VL 30B A3B Thinking31.1B—no speed estimateSpills to system RAMest Cydonia 24B V4.123.6B—no speed estimateSpills to system RAM Uncensored24B—no speed estimateSpills to system RAM Mistral Small 3.2 24B24B—no speed estimateSpills to system RAM Gemma 3 27B27.4B—no speed estimateSpills to system RAM Qwen3.6 27B27.8B—no speed estimateSpills to system RAM Qwen3 VL 30B A3B Instruct31.1B—no speed estimateSpills to system RAMest Qwen3 30B A3B Thinking 250730.5B—no speed estimateSpills to system RAMest Qwen3 Coder 30B A3B Instruct30.5B—no speed estimateSpills to system RAMest Qwen3 30B A3B Instruct 250730.5B—no speed estimateSpills to system RAMest Mistral Small 3.1 24B24B—no speed estimateSpills to system RAM Reka Flash 320.9B—no speed estimateSpills to system RAM Mistral Small 323.6B—no speed estimateSpills to system RAM Gemma 2 27B27.2B—no speed estimateSpills to system RAMest Muse Glimmer 30B29.8B—no speed estimateSpills to system RAMest Nemotron 3.5 Lightning31.6B—no speed estimateSpills to system RAMest Qwen3.8 27B27.8B—no speed estimateSpills to system RAM Hy-MT2-30B-A3B30.1B—no speed estimateSpills to system RAMest Ternary Bonsai 2 27B27B—no speed estimateSpills to system RAMest

Too large for this device101 modelsThe weights do not fit even with part of them offloaded to system memory. Closest calls first.

Qwen3 32B32.8B—no speed estimateToo largeest Qwen3.6 35B A3B36B—no speed estimateToo large Qwen3.5-35B-A3B36B—no speed estimateToo large Skyfall 36B V236.9B—no speed estimateToo large Olmo 3 32B Think32.2B—no speed estimateToo large Qwen3 VL 32B Instruct33.4B—no speed estimateToo large Nex-N2.5-Mini35.1B—no speed estimateToo large Laguna XS 2.133.4B—no speed estimateToo large Gemma 4 31B31.3B—no speed estimateToo large Qwen2.5 Coder 32B Instruct32.8B—no speed estimateToo large Nex-N2-Mini35.1B—no speed estimateToo large Llama 3.3 Euryale 70B70.6B—no speed estimateToo large Qwen2.5 VL 72B Instruct73.4B—no speed estimateToo large Qwen3 Next 80B A3B Thinking81.3B—no speed estimateToo large Qwen3 Coder Next79.7B—no speed estimateToo large Llama 3.3 70B Instruct70.6B—no speed estimateToo large Qwen2.5 72B Instruct72.7B—no speed estimateToo large Qwen3 Next 80B A3B Instruct81.3B—no speed estimateToo large Command A111B—no speed estimateToo large gpt-oss-120b120B—no speed estimateToo large Llama 3.1 70B Instruct70.6B—no speed estimateToo large Hermes 4 70B70.6B—no speed estimateToo large Llama 3.1 Euryale 70B v2.270.6B—no speed estimateToo large R1 Distill Llama 70B70.6B—no speed estimateToo large Hermes 3 70B Instruct70.6B—no speed estimateToo large Qwen3.5-122B-A10B125B—no speed estimateToo large Devstral 2 2512125B—no speed estimateToo large GLM 4.6V108B—no speed estimateToo large Ling-2.6-flash107B—no speed estimateToo large Magnum v4 72B72.7B—no speed estimateToo large Mistral Medium 3.5128B—no speed estimateToo large Laguna S 2.1118B—no speed estimateToo large Hunyuan A13B Instruct80.4B—no speed estimateToo large Mixtral 8x22B Instruct141B—no speed estimateToo large GLM 4.5V108B—no speed estimateToo large Llama 4 Scout109B—no speed estimateToo large GLM 4.5 Air111B—no speed estimateToo large Step 3.7 Flash201B—no speed estimateToo large Mistral Small 4119B—no speed estimateToo large Nemotron 3 Super124B—no speed estimateToo large Ling-3.0-flash128B—no speed estimateToo large MiniMax M2.7229B—no speed estimateToo large MiniMax M2229B—no speed estimateToo large Step 3.5 Flash199B—no speed estimateToo large Qwen3 VL 235B A22B Instruct236B—no speed estimateToo large MiniMax M2.5229B—no speed estimateToo large Qwen3 235B A22B Instruct 2507235B—no speed estimateToo large Qwen3 VL 235B A22B Thinking236B—no speed estimateToo large DeepSeek V4 Flash291B—no speed estimateToo large GLM 5.3 Flash321B—no speed estimateToo large Hy3299B—no speed estimateToo large Laguna M.1226B—no speed estimateToo large Nex-N2-Pro397B—no speed estimateToo large MiniMax M2.1229B—no speed estimateToo large Qwen3 235B A22B Thinking 2507235B—no speed estimateToo large GLM 4.6357B—no speed estimateToo large GLM 4.7358B—no speed estimateToo large GLM 4.5358B—no speed estimateToo large MiniMax M3427B—no speed estimateToo large Inkling Small266B—no speed estimateToo large Trinity Large Thinking399B—no speed estimateToo large Jamba Large 1.7399B—no speed estimateToo large Llama 4 Maverick402B—no speed estimateToo large Hermes 3 405B Instruct406B—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-01456B—no speed estimateToo large MiMo-V2.5311B—no speed estimateToo large Qwen3 Coder 480B A35B480B—no speed estimateToo large DeepSeek V3 0324685B—no speed estimateToo large Nex-N2.5-Pro397B—no speed estimateToo large Qwen3.5 397B A17B403B—no speed estimateToo large Hermes 4 405B406B—no speed estimateToo large GLM 5.1754B—no speed estimateToo large MiniMax M1456B—no speed estimateToo large DeepSeek V3.1 Terminus685B—no speed estimateToo large DeepSeek V3685B—no speed estimateToo large GLM 5754B—no speed estimateToo large Inkling952B—no speed estimateToo large Nemotron 3 Ultra561B—no speed estimateToo large MiMo-V2.5-Pro1T—no speed estimateToo large Kimi K2 07111T—no speed estimateToo large Ling-2.6-1T1T—no speed estimateToo large Kimi K2.7 Code1.1T—no speed estimateToo large R1 0528685B—no speed estimateToo large DeepSeek V3.1685B—no speed estimateToo large DeepSeek V3.2685B—no speed estimateToo large DeepSeek V3.2 Exp685B—no speed estimateToo large Kimi K2 09051T—no speed estimateToo large Ring-2.6-1T1T—no speed estimateToo large Kimi K2.61.1T—no speed estimateToo large Kimi K2.51.1T—no speed estimateToo large GLM 5.2753B—no speed estimateToo large DeepSeek V4.1 Flash763B—no speed estimateToo large Hy4 preview780B—no speed estimateToo large Kimi K2 Thinking1.1T—no speed estimateToo large Qwen3.8 2.4T A95B2.4T—no speed estimateToo large DeepSeek V4 Pro1.6T—no speed estimateToo large LongCat 2.01.8T—no speed estimateToo large Kimi K32.8T—no speed estimateToo large

224 models have a verdict on this device: 98 fit in memory, 25 spill into system memory and 101 are too large. Every one of them is on this page.

03

Other Ada Lovelace cards

The same generation in other memory sizes. Memory is what decides which models fit.

GeForce RTX 4060 8GB8 GB · less memory GeForce RTX 4070 SUPER12 GB · less memory GeForce RTX 4060 Ti 16GB16 GB · same memory GeForce RTX 4080 SUPER16 GB · same memory GeForce RTX 409024 GB · more memory RTX 6000 Ada48 GB · more memory L40S48 GB · more memory

Something wrong on this page? Tell us