FitMyLLM
▸ AMD· RDNA 2.0

AMD Radeon Pro W6800X

VRAM
32 GB
HIGH-END
BANDWIDTH
512
GB/S
MODELS Q4
255/331
77%
7B Q4 SPEED
~65
BLAZING
▸ MODEL COVERAGE @ Q477% OF ALL
▸ ESTIMATED SPEED· BY MODEL SIZE @ Q4

Average speeds at Q4 quantization. Actual performance varies by model architecture and context length.

3B
~152
TOK/S
7B
~65
TOK/S
14B
~33
TOK/S
32B
~14
TOK/S
70B
—
39.4GB NEEDED
▸ SPECIFICATIONS
VRAM
32 GB
BANDWIDTH
512 GB/s
FP16 COMPUTE
32.1 TFLOPS
TDP
200W
MEMORY
GDDR6
ARCHITECTURE
RDNA 2.0
STREAM PROCESSORS
3,840
COMPUTE UNITS
60
205
FAST MODELS · >30 TOK/S
Real-time chat speed
255
USABLE · >10 TOK/S
Comfortable for all tasks
255
TOTAL COMPATIBLE
Fit in VRAM at Q4
▸ DON’T WANT TO BUY?

Test Radeon Pro W6800X (or anything bigger) without committing. Pay by the second, cancel anytime.

Rent on RunPod →Or Vast.ai →

Spin up in ~60s. Pay by the second. Cancel anytime.

Some links are affiliate links — we may earn a small commission at no extra cost to you. This helps keep FitMyLLM free and independent.

▸ COMPATIBLE MODELS· 255
S
nomic-embed-text-v1.5 100M0.1B
EMBEDDING·8K CTX· CHAT
4551
TOK/S · 2% VRAM
›
S
GPT-2 124M0.124B
GPT2·1K CTX· CHAT
3670
TOK/S · 2% VRAM
›
S
SmolLM2 135M0.135B
SMOLLM·2K CTX· CHAT
3371
TOK/S · 2% VRAM
›
S
bge-large-en-v1.5 335M0.335B
EMBEDDING·1K CTX· CHAT
1359
TOK/S · 2% VRAM
›
S
mxbai-embed-large-v10.335B
EMBEDDING·1K CTX· EMBEDDING
1359
TOK/S · 2% VRAM
›
S
Snowflake Arctic Embed L0.335B
EMBEDDING·1K CTX· EMBEDDING
1359
TOK/S · 2% VRAM
›
S
GPT-2 Medium 345M0.345B
GPT2·1K CTX· CHAT
1319
TOK/S · 2% VRAM
›
S
SmolLM2 360M0.36B
SMOLLM·8K CTX· CHAT
1264
TOK/S · 2% VRAM
›
S
Falcon-H1 0.5B0.5B
FALCON·128K CTX· CHAT
910
TOK/S · 2% VRAM
›
S
Qwen 1.5 0.5B0.5B
QWEN·32K CTX· CHAT
910
TOK/S · 2% VRAM
›
S
Qwen 2.5 0.5B0.5B
QWEN·32K CTX· CHAT
910
TOK/S · 2% VRAM
›
S
BGE-M30.568B
EMBEDDING·8K CTX· EMBEDDING
801
TOK/S · 3% VRAM
›
S
Qwen3 0.6B0.6B
QWEN·32K CTX· CHAT· REASONING
759
TOK/S · 3% VRAM
›
S
GPT-2 Large 774M0.774B
GPT2·1K CTX· CHAT
588
TOK/S · 3% VRAM
›
S
Qwen 3.5 0.8B0.8B
QWEN·256K CTX· CHAT· CODING· MULTILINGUAL
569
TOK/S · 3% VRAM
›
S
Qwen3.5-0.8B0.9B
QWEN·256K CTX· CHAT
506
TOK/S · 3% VRAM
›
S
Falcon3-1B1B
FALCON·32K CTX· CHAT
455
TOK/S · 3% VRAM
›
S
InternLM2 1B1B
INTERNLM·32K CTX· CHAT
455
TOK/S · 3% VRAM
›
S
TinyLlama 1.1B1.1B
LLAMA·2K CTX· CHAT
414
TOK/S · 4% VRAM
›
S
LFM2.5-1.2B-Thinking1.2B
LFM·122K CTX· CHAT· REASONING· TOOL_USE
379
TOK/S · 4% VRAM
›
S
Llama-3.2-1B1.2B
LLAMA·4K CTX· CHAT
379
TOK/S · 4% VRAM
›
S
DeepSeek Coder 1.3B1.3B
DEEPSEEK·16K CTX· CODING
350
TOK/S · 4% VRAM
›
S
EXAONE-4.0-1.2B1.3B
EXAONE·64K CTX· CHAT
350
TOK/S · 4% VRAM
›
S
OPT 1.3B1.3B
OPT·2K CTX· CHAT
350
TOK/S · 4% VRAM
›
S
Phi-1 1.3B1.3B
PHI·2K CTX· CODING
350
TOK/S · 4% VRAM
›
S
Phi-1.5 1.3B1.3B
PHI·2K CTX· CHAT· CODING
350
TOK/S · 4% VRAM
›
S
granite-4.0-h-tiny 6.9B6.9BMoE
GRANITE·128K CTX· CHAT
303
TOK/S · 15% VRAM
›
S
Falcon-H1 1.5B1.5B
FALCON·128K CTX· CHAT· CODING
303
TOK/S · 4% VRAM
›
S
GPT-2 XL 1.5B1.5B
GPT2·1K CTX· CHAT
303
TOK/S · 4% VRAM
›
S
Qwen2.5-Coder-1.5B1.5B
QWEN·32K CTX· CHAT· TOOL_USE· CODING
303
TOK/S · 4% VRAM
›
S
Qwen2 Math 1.5B1.5B
QWEN·4K CTX· REASONING
303
TOK/S · 4% VRAM
›
S
Qwen 2.5 1.5B1.5B
QWEN·32K CTX· CHAT· CODING
303
TOK/S · 4% VRAM
›
S
Yi Coder 1.5B1.5B
YI·125K CTX· CODING
303
TOK/S · 4% VRAM
›
S
stablelm-2-1_6b1.6B
STABLELM·4K CTX· CHAT
284
TOK/S · 5% VRAM
›
S
Qwen3 1.7B1.7B
QWEN·32K CTX· CHAT· REASONING
268
TOK/S · 5% VRAM
›
S
SmolLM2 1.7B1.71B
SMOLLM·8K CTX· CHAT
266
TOK/S · 5% VRAM
›
S
Qwen 1.5 1.8B1.8B
QWEN·32K CTX· CHAT
253
TOK/S · 5% VRAM
›
S
Moondream2 1.9B1.9B
OTHER·2K CTX· VISION· CHAT
240
TOK/S · 5% VRAM
›
S
Gemma 1 2B2B
GEMMA·8K CTX· CHAT
228
TOK/S · 5% VRAM
›
S
Granite 3.0 2B2B
GRANITE·128K CTX· CHAT· CODING
228
TOK/S · 5% VRAM
›
▸ NEXT STEP

Get personalized recommendations.

See ranked models with benchmark scores, run commands, and precise speed estimates for your Radeon Pro W6800X.

▸Find models for Radeon Pro W6800XCompare with other GPUs
▸ DEVICE UNDER TEST

AMD Radeon Pro W6800X — 32 GB VRAM.

▸ RADEON PRO W6800X SPEC
BRAND
AMD
VRAM
32 GB GDDR6
BANDWIDTH
512 GB/s
FP16 COMPUTE
32.1 TFLOPS
FP32 COMPUTE
16 TFLOPS
STREAM PROCESSORS
3,840
TDP
200 W
ARCHITECTURE
RDNA 2.0
▸ AI CAPABILITY
255/ 331 models @ Q4

With 32 GB VRAM and 512 GB/s bandwidth, this GPU handles models up to 41.9B parameters.

Speed ≈ bandwidth / model_size × efficiency. A 7B model at Q4 runs at ~59 tok/s.

§ 01TOP MODELS FOR RADEON PRO W6800X
255 FIT · SHOWING 20
MODELSIZEVRAM Q4TOK/SAVG
Phi-3.5 MoE 42B41.9B26.1 GB6956.7
Falcon 40B40B24.9 GB1120.9
Qwen3.5-35B-A3B36B22.5 GB1348.5
c4ai-command-r-v01 35B35B21.9 GB1327.5
Qwen 3.5 35B A3B35B21.9 GB15253.3
Qwen 3.6 35B A3B35B21.9 GB15262.7
Nous Capybara 34B34.4B21.5 GB1342.0
Yi-1.5 34B34.4B21.5 GB1345.3
Falcon-H1 34B34B21.3 GB1366.1
CodeLlama 34B34B21.3 GB1325.4
Nous Hermes 2 34B34B21.3 GB1347.0
Phind CodeLlama 34B34B21.3 GB1368.1
LLaVA-1.6 Yi 34B34B21.3 GB1347.4
WizardCoder Python 34B34B21.3 GB1373.2
Yi 34B34B21.3 GB1333.4
DeepSeek Coder 33B33B20.7 GB1426.0
Vicuna 33B33B20.7 GB1417.2
LLaMA 1 30B33B20.7 GB1417.8
DeepSeek-R1-Distill-Qwen-32B32.8B20.5 GB1446.9
Qwen3 32B32.8B20.5 GB1454.9