What Models?

Pick the right model for your GPU — in seconds.

What AI models can an RTX 4070 SUPER (12 GB) run?

The GeForce RTX 4070 SUPER has 12 GB of VRAM and 504 GB/s memory bandwidth. 37 models run well on it and 3 more are a tight fit.

Fine-tune for your setup (context, speed, system RAM, multiple GPUs) → Check price on Amazon

As an Amazon Associate, I earn from qualifying purchases.

Best for general use

Highest MMLU score among models that run well

  1. Qwen3.5 9B Q8_0 100K context · ~51 tok/s · MMLU 82.5
  2. Qwen 2.5 14B Q4_K_M 16K context · ~50 tok/s · MMLU 79.9
  3. Qwen3.5 4B Q8_0 252K context · ~97 tok/s · MMLU 79.1

Best for coding

Highest SWE-bench Verified score among models that run well

  1. K2-Horizon 7B Q8_0 22K context · ~51 tok/s · SWE-bench 70.6
  2. Ornith 1.5 9B Q8_0 93K context · ~50 tok/s · SWE-bench 70.6
  3. MiMo V2.6 Distill Qwen 9B Q8_0 100K context · ~51 tok/s · SWE-bench 61.1

Upgrade options

The cheapest GPUs with more VRAM than the RTX 4070 SUPER, by launch MSRP.

As an Amazon Associate, I earn from qualifying purchases.

Every model on the RTX 4070 SUPER

72 of 200 model configurations can run on 12 GB — sorted by quality

100 models, each counted once by the best way it runs on your hardware.

Largest that runs well StarCoder2 15B 15.6B params · Q4_K_M · 16K context
Fastest good-quality model Phi-3 Mini 3.8B ~149 tok/s · Q4_K_M · 26K context
? MMLU (Massive Multitask Language Understanding) measures general knowledge across 57 subjects. Higher = more capable. Sorted best-first.

Runs well 37

  • Qwen3.5 9B 9B Great VisionTool useReasoning
    MMLU 82.5
    Q4_K_M
    Context 216K Speed ~80 tok/s
    Q8_0
    Context 100K Speed ~51 tok/s
  • Qwen 2.5 14B 14.7B Great Tool use
    MMLU 79.9
    Q4_K_M
    Context 16K Speed ~50 tok/s
    Q8_0 Doesn't fit · needs 15.9 GB
  • Qwen3.5 4B 4B Great VisionTool useReasoning
    MMLU 79.1
    Q4_K_M
    Context 256K Speed ~142 tok/s
    Q8_0
    Context 252K Speed ~97 tok/s

Runs well 37

  • Qwen3.5 9B 9B Great VisionTool useReasoning
    MMLU 82.5
    Q4_K_M
    Context 216K Speed ~80 tok/s
    Q8_0
    Context 100K Speed ~51 tok/s
  • Qwen 2.5 14B 14.7B Great Tool use
    MMLU 79.9
    Q4_K_M
    Context 16K Speed ~50 tok/s
    Q8_0 Doesn't fit · needs 15.9 GB
  • Qwen3.5 4B 4B Great VisionTool useReasoning
    MMLU 79.1
    Q4_K_M
    Context 256K Speed ~142 tok/s
    Q8_0
    Context 252K Speed ~97 tok/s
  • Qwen3 8B 8.2B Great ReasoningTool use
    MMLU 79
    Q4_K_M
    Context 50K Speed ~84 tok/s
    Q8_0
    Context 24K Speed ~52 tok/s
  • Gemma 4 12B 11.96B Great VisionTool useReasoning
    MMLU 77.2
    Q4_K_M
    Context 256K Speed ~66 tok/s
    Q8_0 Tight fit · 13K context
  • Nemotron Nano 12B 12.6B Great ReasoningTool use
    MMLU 76
    Q4_K_M
    Context 128K Speed ~60 tok/s
    Q8_0 Doesn't fit · needs 13.1 GB
  • Qwen 2.5 7B 7.62B Good Tool use
    MMLU 74.2
    Q4_K_M
    Context 128K Speed ~89 tok/s
    Q8_0
    Context 73K Speed ~55 tok/s
  • Qwen 2.5 VL 7B 7.62B Good VisionTool use
    MMLU 74.2
    Q4_K_M
    Context 128K Speed ~89 tok/s
    Q8_0
    Context 73K Speed ~55 tok/s
  • Gemma 3 12B 11.8B Good Vision
    MMLU 74
    Q4_K_M
    Context 77K Speed ~61 tok/s
    Q8_0 Doesn't fit · needs 12.6 GB
  • Qwen 2.5 Coder 14B 14.7B Good
    MMLU 74
    Q4_K_M
    Context 16K Speed ~50 tok/s
    Q8_0 Doesn't fit · needs 15.9 GB
  • Llama 3.2 Vision 11B 10.6B Good VisionTool use
    MMLU 73
    Q4_K_M
    Context 49K Speed ~72 tok/s
    Q8_0 Tight fit · 13K context
  • DeepSeek R1 Distill Qwen 14B 14.7B Good Reasoning
    MMLU 72
    Q4_K_M
    Context 16K Speed ~50 tok/s
    Q8_0 Doesn't fit · needs 15.9 GB
  • Phi-3 Mini 3.8B 3.82B Good
    MMLU 70.9
    Q4_K_M
    Context 26K Speed ~149 tok/s
    Q8_0
    Context 21K Speed ~100 tok/s
  • Phi-4 Mini 3.8B 3.84B Good
    MMLU 70
    Q4_K_M
    Context 77K Speed ~144 tok/s
    Q8_0
    Context 64K Speed ~99 tok/s
  • Gemma 4 E4B 4.5B Good VisionTool useReasoning
    MMLU 69.4
    Q4_K_M
    Context 128K Speed ~84 tok/s
    Q8_0
    Context 128K Speed ~59 tok/s
  • Llama 3.1 8B 8.03B Good Tool use
    MMLU 69.4
    Q4_K_M
    Context 58K Speed ~85 tok/s
    Q8_0
    Context 28K Speed ~53 tok/s
    fp16 Doesn't fit · needs 16.2 GB
  • Qwen 2.5 Coder 7B 7.62B Good
    MMLU 68
    Q4_K_M
    Context 128K Speed ~89 tok/s
    Q8_0
    Context 73K Speed ~55 tok/s
  • Mistral Nemo 12B 12.2B Good Tool use
    MMLU 68
    Q4_K_M
    Context 23K Speed ~59 tok/s
    Q8_0 Doesn't fit · needs 13.3 GB
  • Qwen 2.5 VL 3B 3.09B Fair Vision
    MMLU 65.6
    Q4_K_M
    Context 128K Speed ~172 tok/s
    Q8_0
    Context 128K Speed ~117 tok/s
  • Qwen 2.5 3B 3.09B Fair Tool use
    MMLU 65.6
    Q4_K_M
    Context 32K Speed ~172 tok/s
    Q8_0
    Context 32K Speed ~117 tok/s
  • Ministral 8B 8.02B Fair Tool use
    MMLU 65
    Q4_K_M
    Context 50K Speed ~83 tok/s
    Q8_0
    Context 25K Speed ~53 tok/s
  • SmallThinker 3B 3.09B Fair Reasoning
    MMLU 65
    Q4_K_M
    Context 32K Speed ~172 tok/s
    Q8_0
    Context 32K Speed ~117 tok/s
  • Llama 3.2 3B 3.21B Fair Tool use
    MMLU 63.4
    Q4_K_M
    Context 93K Speed ~167 tok/s
    Q8_0
    Context 80K Speed ~114 tok/s
    fp16
    Context 52K Speed ~68 tok/s
  • Gemma 3 4B 3.88B Fair Vision
    MMLU 63
    Q4_K_M
    Context 128K Speed ~144 tok/s
    Q8_0
    Context 128K Speed ~98 tok/s
  • Gemma 4 E2B 2.3B Fair VisionTool useReasoning
    MMLU 60
    Q4_K_M
    Context 128K Speed ~119 tok/s
    Q8_0
    Context 128K Speed ~90 tok/s
  • DeepSeek R1 Distill Qwen 7B 7.62B Fair Reasoning
    MMLU 58
    Q4_K_M
    Context 128K Speed ~89 tok/s
    Q8_0
    Context 73K Speed ~55 tok/s
  • Qwen3.5 2B 2B Fair VisionTool useReasoning
    MMLU 55.3
    Q4_K_M
    Context 256K Speed ~230 tok/s
    Q8_0
    Context 256K Speed ~176 tok/s
  • StarCoder2 15B 15.6B Basic
    MMLU 51.4
    Q4_K_M
    Context 16K Speed ~46 tok/s
    Q8_0 Doesn't fit · needs 17.1 GB
  • DeepSeek Coder 6.7B 6.74B Basic
    MMLU 50
    Q4_K_M
    Context 16K Speed ~99 tok/s
    Q8_0 Tight fit · 9K context
  • Llama 3.2 1B 1.24B Basic Tool use
    MMLU 49.3
    Q4_K_M
    Context 128K Speed ~278 tok/s
    Q8_0
    Context 128K Speed ~217 tok/s
  • Qwen3.5 0.8B 0.8B Basic VisionTool useReasoning
    MMLU 29.7
    Q4_K_M
    Context 256K Speed ~336 tok/s
    Q8_0
    Context 256K Speed ~286 tok/s
  • Nemotron 3 Nano 4B 3.97B Basic ReasoningTool use
    MMLU
    Q4_K_M
    Context 256K Speed ~134 tok/s
    Q8_0
    Context 256K Speed ~102 tok/s
  • Ornith 1.5 9B 9.65B Basic Tool useReasoning
    MMLU
    Q4_K_M
    Context 213K Speed ~79 tok/s
    Q8_0
    Context 93K Speed ~50 tok/s
  • MiMo V2.6 Distill Qwen 9B 9.41B Basic VisionTool useReasoning
    MMLU
    Q4_K_M
    Context 211K Speed ~78 tok/s
    Q8_0
    Context 100K Speed ~51 tok/s
  • LFM2.5 2.6B 2.7B Basic Tool use
    MMLU
    Q4_K_M
    Context 128K Speed ~197 tok/s
    Q8_0
    Context 128K Speed ~137 tok/s
  • Nemotron 3.5 Bolt E4B 8B Basic Tool useReasoning
    MMLU
    Q4_K_M
    Context 64K Speed ~75 tok/s
    Q8_0
    Context 64K Speed ~59 tok/s
  • K2-Horizon 7B 9B Basic Tool useReasoning
    MMLU
    Q4_K_M
    Context 49K Speed ~81 tok/s
    Q8_0
    Context 22K Speed ~51 tok/s

Tight fit 3

Notify me when new models fit my GPU

An email when a new model fits your RTX 4070 SUPER — no spam, unsubscribe anytime.

Doesn't fit 60

  • DeepSeek R1 0528 671B Excellent Reasoning
    MMLU 90.8
    Q4_K_M Needs 409.3 GB — 397.3 GB over
    Q8_0 Needs 713.4 GB — 701.4 GB over
  • Qwen3 Next 80B-A3B 80B Excellent Tool use
    MMLU 90
    Q4_K_M Needs 48.4 GB — 36.4 GB over
    Q8_0 Needs 84.8 GB — 72.8 GB over
    GPUs that can run this

    As an Amazon Associate, I earn from qualifying purchases.

  • MiMo V2.5 Pro 1020B Excellent Reasoning
    MMLU 89.4
    Q4_K_M Needs 630.1 GB — 618.1 GB over
    Q8_0 Needs 1090.1 GB — 1078.1 GB over

Doesn't fit 60