Will It Run AI
CalculatorModelsHardwareCompare
Product
  • Calculator
  • Compare
  • Tier List
Browse
  • Models
  • Hardware
  • Docs
About
  • Why It Works
  • What's New
  • Legal Notice
  • Privacy Policy

All estimates are approximations based on mathematical models and public specifications. Actual performance may vary. Do not make purchasing decisions based solely on these estimates.

Data sourced from Hugging Face, Ollama, and official model documentation. Model names and logos are trademarks of their respective owners.

© 2026 Will It Run AI — Fase Consulting Ibiza, S.L. (NIF: B57969656)

Home/Hardware/GPUs/RX 6750 XT 12GB
AMD

AMD

RX 6750 XT 12GB

RX 6000ConsumerRDNA 2PCIe 4ROCm
12GB
VRAM
432GB/s
Bandwidth
30TFLOPS
FP16 Compute
240TOPS
INT8 Inference
VRAM12 GBBandwidth432 GB/sCompute30 TFInference240 TOPS
RX 6750 XT 12GBCategory AvgMacBook Pro M3 Pro 18GB

Specifications

Compute
FP1630 TFLOPS
INT8240 TOPS
ArchitectureRDNA 2
Memory
VRAM12 GB
Bandwidth432 GB/s
General
FamilyRX 6000
SegmentConsumer
InterconnectPCIe 4
Compute PlatformROCM

Architecture

RDNA 2

RDNA 2 is AMD's second-generation RDNA architecture, built on TSMC 7nm. It introduced hardware ray tracing and Infinity Cache for improved bandwidth efficiency. Powers the RX 6000 series and is also used in gaming consoles.

AI Relevance

Limited official ROCm support for consumer RDNA 2 cards — most AI runtimes require workarounds. Can run smaller models via llama.cpp with Vulkan or HIP backends, but performance is well behind NVIDIA equivalents.

Process: TSMC 7nmPlatform: ROCMPrecisions: FP32, FP16, INT8

Recommendations by Workload

Agentic Coding

C

Granite 3.1 8B

This model is still usable for agentic-coding, but it is not the most specialized pick. It sits in the middle of the current model mix. It fits natively with comfortable headroom. Known channels: huggingface, ollama.

Decode 46.9 tok/s · 41K ctx · llama.cpp
9.5 GB / 12.0 GB VRAM

Chat

C

Qwen 3 8B

This model is a direct match for chat. It belongs to a current frontier family for local AI. It fits natively with comfortable headroom. Known channels: huggingface, ollama, lm-studio.

Decode 46.9 tok/s · 12K ctx · llama.cpp
7.8 GB / 12.0 GB VRAM

Coding

C

Codestral Mamba 7B

This model is still usable for coding, but it is not the most specialized pick. It sits in the middle of the current model mix. It fits natively with comfortable headroom. Known channels: huggingface, ollama.

Decode 53.6 tok/s · 26K ctx · llama.cpp
7.5 GB / 12.0 GB VRAM

RAG

C

granite 8b code instruct 4k

This model is a direct match for rag. It sits in the middle of the current model mix. It fits natively with comfortable headroom.

Decode 46.9 tok/s · 41K ctx · llama.cpp
9.5 GB / 12.0 GB VRAM

Reasoning

C

Qwen 3 8B

This model is a direct match for reasoning. It belongs to a current frontier family for local AI. It fits natively with comfortable headroom. Known channels: huggingface, ollama, lm-studio.

Decode 46.9 tok/s · 23K ctx · llama.cpp
8.2 GB / 12.0 GB VRAM

Full Model Compatibility

UnslothQwen3.5 9B
C55
9B9.0 GB42 tok/s21K ctx
dense
BartowskiBMeta Llama 3.1 8B Instruct
C55
8B8.2 GB47 tok/s23K ctx
dense
HauhauCSHQwen3.5 9B Uncensored HauhauCS Aggressive
C55
9B9.0 GB42 tok/s21K ctx
dense
XtunerXllava llama 3 8b v1 1
C55
8B8.2 GB47 tok/s23K ctx
dense
UnslothDeepSeek R1 0528 Qwen3 8B
C55
8B8.2 GB47 tok/s23K ctx
dense
MaziyarPanahiMMeta Llama 3 8B Instruct
C55
8B8.2 GB47 tok/s23K ctx
dense
Lmstudio-communityLQwen3.5 9B
C55
9B9.0 GB42 tok/s21K ctx
dense
TheBlokeTLlama 2 7B Chat
C54
7B7.5 GB54 tok/s26K ctx
dense
TheBlokeTMistral 7B Instruct v0.2
C54
7B7.5 GB54 tok/s26K ctx
dense
MaziyarPanahiMMistral 7B Instruct v0.3
C54
7B7.5 GB54 tok/s26K ctx
dense
UnslothQwen3.5 4B
C52
4B5.3 GB94 tok/s36K ctx
dense
Lmstudio-communityLgemma 3 4b it
C52
4B5.3 GB94 tok/s36K ctx
dense
BartowskiBLlama 3.2 3B Instruct
C51
3B5.1 GB108 tok/s38K ctx
dense
QwenQwen2.5 3B Instruct
C51
3B4.7 GB125 tok/s41K ctx
dense
BartowskiBgemma 2 2b it
C50
2B4.5 GB147 tok/s42K ctx
dense
Googlegemma 2b
C50
2B4.1 GB188 tok/s47K ctx
dense
TheDrummerTGemmasutra Mini 2B v1
C49
2B4.1 GB188 tok/s47K ctx
dense
QwenQwen2.5 1.5B Instruct
C49
1.5B3.8 GB229 tok/s50K ctx
dense
Hugging-quantsHLlama 3.2 1B Instruct Q8 0
C49
1B3.7 GB240 tok/s52K ctx
dense
TheBlokeTTinyLlama 1.1B Chat v1.0
C49
1.1B3.6 GB229 tok/s54K ctx
dense
Ggml-orgGSmolVLM 500M Instruct
C48
0.5B3.3 GB240 tok/s58K ctx
dense
Ggml-orgGembeddinggemma 300M
C48
0.3B3.1 GB240 tok/s61K ctx
dense
DeepSeekDeepSeek R1 671B
F0
671B417.2 GB2 tok/s4K ctx
moe
MistralDevstral 2 123B Instruct
F0
123B96.3 GB3 tok/s4K ctx
dense
Z.aiGLM-5
F0
744B462.2 GB2 tok/s4K ctx
moe
UnslothQwen3.5 27B
F0
27B22.8 GB14 tok/s8K ctx
dense
UnslothQwen3.5 35B A3B
F0
35B28.9 GB11 tok/s7K ctx
dense
Moonshot AIKimi K2.5
F0
1000B617.1 GB2 tok/s4K ctx
moe
MistralMistral Large 3
F0
675B420.3 GB2 tok/s4K ctx
+1moe
MistralMistral Small 4 119B
F0
119B75.7 GB9 tok/s4K ctx
moe
AlibabaQwen3-Coder 30B A3B Instruct
F0
30.5B21.5 GB32 tok/s9K ctx
moe
AlibabaQwen3-Coder 480B A35B Instruct
F0
480B300.4 GB2 tok/s4K ctx
moe
AlibabaQwen3-Coder-Next
F0
80B51.7 GB14 tok/s4K ctx
moe
UnslothQwen3.5 122B A10B
F0
122B80.9 GB4 tok/s4K ctx
dense
DeepSeekDeepSeek V3 671B
F0
671B417.2 GB2 tok/s4K ctx
moe
MistralMixtral 8x22B
F0
141B94.2 GB5 tok/s4K ctx
moe
AlibabaQwen 2.5 72B
F0
72B57.3 GB5 tok/s4K ctx
dense
AlibabaQwen 3 235B A22B
F0
235B148.9 GB4 tok/s4K ctx
moe
AlibabaQwen3-VL 30B A3B Instruct
F0
30B21.2 GB33 tok/s9K ctx
moe
UnslothQwen3.5 397B A17B
F0
397B306.3 GB2 tok/s4K ctx
dense
MistralDevstral Small 2 24B Instruct
F0
24B20.5 GB16 tok/s9K ctx
dense
MetaLlama 3.3 70B
F0
70B55.7 GB5 tok/s4K ctx
dense
MetaLlama 4 Maverick 17B 128E
F0
400B248.8 GB3 tok/s4K ctx
moe
CohereCommand A 111B
F0
111B87.2 GB3 tok/s4K ctx
dense
AlibabaQwen 2.5 Coder 32B
F0
32B26.6 GB12 tok/s7K ctx
dense
AlibabaQwen 2.5 VL 72B
F0
72B57.3 GB5 tok/s4K ctx
dense
Unslothgemma 3 27b it
F0
27B22.8 GB14 tok/s8K ctx
dense
Lmstudio-communityLQwen3.5 35B A3B
F0
35B28.9 GB11 tok/s7K ctx
dense
MistralCodestral 2 25.08
F0
22B19.0 GB17 tok/s10K ctx
dense
MistralDevstral Small 1.1
F0
24B20.5 GB16 tok/s9K ctx
dense

Just out of reach

Models you could run with an upgrade

High-quality models that need a bit more memory

DeepSeekDeepSeek R1 671B
671BTier 5Needs ~423.0 GB
MistralDevstral 2 123B Instruct
123BTier 5Needs ~115.6 GB
Runs on Mac Studio M3 Ultra 256GB
Z.aiGLM-5
744BTier 5Needs ~468.4 GB
UnslothQwen3.5 27B
27BTier 5Needs ~27.0 GB
Runs on RTX 5090 32GB (~$1,999)
UnslothQwen3.5 35B A3B
35BTier 5Needs ~34.4 GB
Runs on Mac mini M4 64GB (~$1,099)

Upgrade paths

Upgrade from RX 6750 XT 12GB

See what you unlock with more powerful hardware

Upgrade options

Upgrade options

AppleMacBook Pro M3 Pro 18GBNext step up
18 GB Unified (+6)
B
Unlocks StableLM 2 12B

~$1,999 MSRP

AMDRX 7600 XT 16GBBest value
16 GB VRAM (+4)
A
Unlocks StarCoder 15B, DeepSeek R1 Distill Qwen 14B, Phi-4-reasoning-plus 14B+23 more

~$329 MSRP

AMDRX 6800 16GBAMD upgrade
16 GB VRAM (+4)512 GB/s (+80)
A
Unlocks StarCoder 15B, DeepSeek R1 Distill Qwen 14B, Phi-4-reasoning-plus 14B+23 more · +14% faster avg

 

AMDAMD Instinct MI350X 288GBBiggest leap
288 GB VRAM (+276)8000 GB/s (+7568)
A
Unlocks Devstral 2 123B Instruct, Qwen3.5 27B, Qwen3.5 35B A3B+107 more · +1728% faster avg

~$8,000 MSRP

Compare this GPU