What can the RTX 4070 Ti Super run?
| Text models: 247 | Visual models: 18 | Strongest first |
- Memory
- 16 GB
- Bandwidth
- 672 GB/s
- Maker
- NVIDIA
Open the calculator on this machineEvery model for 16 GB of VRAM
Text models that run on it
| Model | Best fit | Context | Speed | Strongest at |
|---|---|---|---|---|
| Qwen3.8 27B | 3-bit | 20K | 34 tok/s | Chat, Coding, Agents, Docs |
| Muse Glimmer 30B | 3-bit | 40K | 32 tok/s | Docs |
| Gemma 4 26B A4B | 3-bit | 224K | 174 tok/s | Chat, Coding, Agents, Docs |
| Qwen3.5 9B | 8-bit | 144K | 49 tok/s | Chat, Coding, Agents, Docs |
| Gemma 4 12B | 8-bit | 248K | 35 tok/s | Chat, Coding, Docs |
| Qwen3.5 4B | 8-bit | 256K | 86 tok/s | Agents, Docs |
| Devstral Small 2 24B | 3-bit | 20K | 39 tok/s | Coding |
| LFM2.5 24B A2B | 4-bit | 44K | 201 tok/s | Agents |
| Mistral Small 3.2 24B | 3-bit | 20K | 39 tok/s | Chat, Coding, Agents |
| Ministral 3 14B | 6-bit | 20K | 41 tok/s | Coding, Agents, Docs |
| Gemma 4 E4B | 8-bit | 128K | 77 tok/s | Chat, Coding, Docs |
| gpt-oss 20B | MXFP4 | 120K | 145 tok/s | Agents |
| Ministral 3 8B | 8-bit | 40K | 50 tok/s | Docs |
| Gemma 4 E2B | 8-bit | 128K | 137 tok/s | Chat |
| Nemotron 3 Nano 4B | 8-bit | 256K | 106 tok/s | Docs |
| Llama 3.1 8B | 8-bit | 48K | 56 tok/s | Chat, Coding, Agents, Docs |
| Qwen3 0.6B | 8-bit | 40K | 394 tok/s | Not rated |
| Qwen3 14B | 6-bit | 16K | 39 tok/s | Not rated |
| Qwen3 VL 8B Instruct | 8-bit | 40K | 51 tok/s | Text and images |
| Qwen3 8B | 8-bit | 40K | 55 tok/s | Not rated |
| Qwen2.5 7B Instruct | 8-bit | 32K | 59 tok/s | Not rated |
| Qwen2.5 VL 7B Instruct | 8-bit | 112K | 55 tok/s | Text and images |
| Unlimited OCR | 8-bit | 32K | 303 tok/s | Reading documents |
| Qwen3 VL 4B Instruct | 8-bit | 64K | 90 tok/s | Text and images |
| Qwen3 1.7B | 8-bit | 40K | 184 tok/s | Not rated |
| Qwen3 4B | 8-bit | 40K | 98 tok/s | Not rated |
| Qwen3 4B Instruct 2507 | 8-bit | 72K | 98 tok/s | Not rated |
| Qwen2.5 3B Instruct | 8-bit | 32K | 124 tok/s | Not rated |
| Qwen3.6 27B | 3-bit | 20K | 34 tok/s | Text and images |
| DeepSeek OCR | 8-bit | 8K | 303 tok/s | Reading documents |
| Qwen2.5 Coder 14B Instruct | 6-bit | 16K | 39 tok/s | Coding |
| Qwen2.5 VL 3B Instruct | 8-bit | 125K | 105 tok/s | Text and images |
| Qwen3.5 27B | 3-bit | 20K | 34 tok/s | Text and images |
| Qwen2.5 Coder 7B Instruct | 8-bit | 32K | 59 tok/s | Coding |
| Mistral 7B Instruct V0.2 | 8-bit | 32K | 59 tok/s | Not rated |
| Chandra OCR 2 | 8-bit | 256K | 77 tok/s | Reading documents |
| Qwen3 VL 2B Instruct | 8-bit | 112K | 177 tok/s | Text and images |
| MiniCPM5 2B | 8-bit | 128K | 158 tok/s | Not rated |
| Gemma 3 4B IT | 8-bit | 128K | 109 tok/s | Text and images |
| Llama 3.2 3B Instruct | 8-bit | 96K | 124 tok/s | Not rated |
| Qwen2.5 14B Instruct | 6-bit | 16K | 39 tok/s | Not rated |
| Qwen3 1.7B Base | 8-bit | 32K | 212 tok/s | Not rated |
| Diffusiongemma 26B A4B IT | 3-bit | 112K | 168 tok/s | Text and images |
| HunyuanOCR | 8-bit | 128K | 308 tok/s | Text and images |
| Qwen3 0.6B Base | 8-bit | 32K | 457 tok/s | Not rated |
| Qwen2 VL 7B Instruct | 8-bit | 32K | 55 tok/s | Text and images |
| Ornith 1.5 9B | 8-bit | 144K | 49 tok/s | Not rated |
| Unlimited OCR 2 | 8-bit | 128K | 302 tok/s | Reading documents |
| Granite 4.1 3B | 8-bit | 128K | 113 tok/s | Not rated |
| Mistral 7B V0.1 | 8-bit | 32K | 59 tok/s | Not rated |
| Qwen2.5 Coder 3B Instruct | 8-bit | 32K | 124 tok/s | Coding |
| Meta Llama 3 8B Instruct Abliterated V3 | 8-bit | 8K | 56 tok/s | Not rated |
| EXAONE 3.5 7.8B Instruct | 8-bit | 32K | 57 tok/s | Not rated |
| Saiga Llama3 8B | 8-bit | 8K | 56 tok/s | Not rated |
| DeepSeek R1 0528 Qwen3 8B | 8-bit | 44K | 55 tok/s | Not rated |
| Gemma 3 12B IT | 8-bit | 28K | 37 tok/s | Text and images |
| Phi 3.5 Mini Instruct | 8-bit | 28K | 107 tok/s | Not rated |
| Qwen3 4B Thinking 2507 | 8-bit | 72K | 98 tok/s | Not rated |
| Phi 4 Mini Instruct | 8-bit | 80K | 104 tok/s | Not rated |
| Dots.ocr | 8-bit | 128K | 139 tok/s | Reading documents |
Best fit is the highest quality, up to 8-bit, that leaves the use case’s context. Ranked by our use-case scores, then by Hugging Face downloads. Speeds are rough, plus or minus 25%. The top 60 of 247 are shown.
Image and video models
| Model | Best fit | Per output | Strongest at |
|---|---|---|---|
| LTX-2.5 | 5-bit | 1.6 min | Video |
| Z-Image Turbo | 16-bit | 11 s | Images |
| FLUX.1 [dev] | 8-bit | 21 s | Images |
| Qwen-Image | 5-bit | 7.5 min | Images |
| Qwen-Image 2.1 | 16-bit | 2.8 min | Images |
| FLUX.3 Klein | 16-bit | 16 s | Images |
| Wan2.2 I2V A14B | 4-bit | 66 min | Video |
| FLUX.2 [klein] 9B | 8-bit | 5.7 s | Images |
| Wan2.1 I2V 14B | 4-bit | 66 min | Video |
| Wan 2.1 T2V-14B | 4-bit | 83 min | Video |
| Krea 2 Turbo | 8-bit | 21 s | Images |
| Stable Diffusion 3.5 Large | 8-bit | 89 s | Images |
| Qwen Image 2512 | 5-bit | 7.5 min | Images |
| MOSS-Image | 16-bit | 1.8 min | Images |
| Wan 2.2 TI2V-5B | 16-bit | 16 min | Video |
| MOSS Image Base | 16-bit | 1.8 min | Images |
| Wan 2.2 A14B | 4-bit | 66 min | Video |
| HunyuanVideo 1.5 | 8-bit | 71 min | Video |
Only just too big
- Qwen3.6 35B A3B
- Gemma 4 31B
- GLM-4.7 Flash
- Nemotron 3.5 Lightning 30B A3B
- Olmo 3.1 32B
- MiniMax H3
- Qwen3 32B
- Qwen3 Coder 30B A3B Instruct
- Qwen3.5 35B A3B
- NVIDIA Nemotron 3 Nano 30B A3B BF16
- Qwen3 30B A3B
- Qwen3 30B A3B Instruct 2507
- Laguna XS.2
- OTel 2.0 LLM 31B IT
- Qwen2.5 32B Instruct
- MiniCPM V 4.7 35B A3B
- Laguna XS 2.1
- Qwen2.5 VL 32B Instruct
- Qwen2.5 Coder 32B Instruct
- Qwen3 VL 30B A3B Instruct
- DeepSeek R1 Distill Qwen 32B
- Gemma 3 27B IT
- FLUX.2 [dev]
- Ornith 1.5 35B A3B
- Qwen3 VL 32B Instruct
- Qwen3 30B A3B Thinking 2507
- Gnani Evon V3.3 30B A3B
- EXAONE 3.5 32B Instruct
- Flint
- Qwen1.5 32B Chat
These need up to a quarter more memory than the RTX 4070 Ti Super has. Spilling into system RAM or a lower precision can get them running, slowly.
Questions
What is the best LLM for the RTX 4070 Ti Super?
Of the open models we track, the strongest that run well on the RTX 4070 Ti Super are Qwen3.8 27B, Muse Glimmer 30B, Gemma 4 26B A4B. 247 text models fit it in all.
Calculated from published specs, not benchmarked. All machines