FitMyLLM
AMD· GCN 1.0

AMD FirePro S7000

Running LLMs on the FirePro S7000 — the long read: which models fit at which quantisation, and the settings worth changing. · Or what a budget buys

VRAM
4 GB
BUDGET
BANDWIDTH
153.6
GB/S
MODELS Q4
97/449
22%
7B Q4 SPEED
~20
GOOD
▸ MODEL COVERAGE @ Q422% OF ALL
▸ ESTIMATED SPEED· BY MODEL SIZE @ Q4

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

3B
~46
TOK/S
7B
3.9GB NEEDED
14B
7.9GB NEEDED
32B
18.0GB NEEDED
70B
39.4GB NEEDED
▸ MEASURED RIG REPORTS

We rent the machine and time every model on it: decode, VRAM peak, concurrency, watts and cost per million tokens. Yours may already be one of them — and two are free to read in full.

SEE THE REPORTS
▸ SPECIFICATIONS
VRAM
4 GB
BANDWIDTH
153.6 GB/s
FP16 COMPUTE
2.4 TFLOPS
TDP
150W
MEMORY
GDDR5
ARCHITECTURE
GCN 1.0
STREAM PROCESSORS
1,280
PCIE
Gen 3 x16
97
FAST MODELS · >30 TOK/S
Real-time chat speed
97
USABLE · >10 TOK/S
Comfortable for all tasks
97
TOTAL COMPATIBLE
Fit in VRAM at Q4
▸ DON’T WANT TO BUY?

Test FirePro S7000 (or anything bigger) without committing. Pay by the second, cancel anytime.

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· 97
S
Falcon-H1R Tiny 90M0.09B
FALCON·256K CTX· CHAT· REASONING
1517
TOK/S · 14% VRAM
S
SmolLM2 135M0.135B
SMOLLM·2K CTX· CHAT
1011
TOK/S · 14% VRAM
S
nomic-embed-text-v1.5 100M0.14B
EMBEDDING·8K CTX· CHAT
975
TOK/S · 14% VRAM
S
GPT-2 124M0.14B
GPT2·1K CTX· CHAT
975
TOK/S · 14% VRAM
S
SmolVLM 256M0.256B
SMOLLM·8K CTX· CHAT· VISION
533
TOK/S · 16% VRAM
S
Gemma 3 270M0.27B
GEMMA·32K CTX· CHAT
506
TOK/S · 16% VRAM
S
Snowflake Arctic Embed M v2.00.305B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
448
TOK/S · 17% VRAM
S
bge-large-en-v1.5 335M0.335B
EMBEDDING·1K CTX· CHAT
408
TOK/S · 17% VRAM
S
mxbai-embed-large-v10.335B
EMBEDDING·1K CTX· EMBEDDING
408
TOK/S · 17% VRAM
S
Snowflake Arctic Embed L0.335B
EMBEDDING·1K CTX· EMBEDDING
408
TOK/S · 17% VRAM
S
LFM2 350M0.35B
LFM·125K CTX· CHAT· TOOL_USE
390
TOK/S · 18% VRAM
S
SmolLM2 360M0.36B
SMOLLM·8K CTX· CHAT
379
TOK/S · 18% VRAM
S
GPT-2 Medium 345M0.38B
GPT2·1K CTX· CHAT
359
TOK/S · 18% VRAM
S
Qwen 2.5 0.5B0.5B
QWEN·32K CTX· CHAT
273
TOK/S · 20% VRAM
S
SmolVLM 500M0.5B
SMOLLM·8K CTX· CHAT· VISION
273
TOK/S · 20% VRAM
S
Falcon-H1 0.5B0.52B
FALCON·128K CTX· CHAT
263
TOK/S · 20% VRAM
S
BGE-M30.568B
EMBEDDING·8K CTX· EMBEDDING
240
TOK/S · 21% VRAM
S
Snowflake Arctic Embed L v2.00.568B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
240
TOK/S · 21% VRAM
S
Qwen3-Embedding 0.6B0.6B
EMBEDDING·32K CTX· EMBEDDING· MULTILINGUAL
228
TOK/S · 21% VRAM
S
Falcon-H1R Tiny 0.6B0.6B
FALCON·32K CTX· CHAT· REASONING
228
TOK/S · 21% VRAM
S
Falcon Perception 0.6B0.6B
FALCON·4K CTX· VISION
228
TOK/S · 21% VRAM
S
Qwen 1.5 0.5B0.62B
QWEN·32K CTX· CHAT
220
TOK/S · 22% VRAM
S
LFM2 700M0.74B
LFM·125K CTX· CHAT· TOOL_USE
185
TOK/S · 24% VRAM
S
Qwen3 0.6B0.75B
QWEN·32K CTX· CHAT· REASONING
182
TOK/S · 24% VRAM
S
GPT-2 Large 774M0.81B
GPT2·1K CTX· CHAT
169
TOK/S · 25% VRAM
S
Qwen 3.5 0.8B0.87B
QWEN·256K CTX· CHAT· CODING· MULTILINGUAL
157
TOK/S · 26% VRAM
S
InternVL3 1B0.94B
OTHER·32K CTX· CHAT· VISION
145
TOK/S · 27% VRAM
S
MiniCPM5 1B1.08B
MINICPM·128K CTX· CHAT· REASONING· MULTILINGUAL
126
TOK/S · 29% VRAM
S
TinyLlama 1.1B1.1B
LLAMA·2K CTX· CHAT
124
TOK/S · 29% VRAM
S
LFM2.5-1.2B-Thinking1.2B
LFM·122K CTX· CHAT· REASONING· TOOL_USE
114
TOK/S · 31% VRAM
S
Llama-3.2-1B1.2B
LLAMA·4K CTX· CHAT
114
TOK/S · 31% VRAM
S
LFM2 1.2B1.2B
LFM·125K CTX· CHAT· TOOL_USE· MULTILINGUAL
114
TOK/S · 31% VRAM
S
Zamba2 1.2B1.2B
OTHER·4K CTX· CHAT
114
TOK/S · 31% VRAM
S
EXAONE-4.0-1.2B1.3B
EXAONE·64K CTX· CHAT
105
TOK/S · 32% VRAM
S
OPT 1.3B1.3B
OPT·2K CTX· CHAT
105
TOK/S · 32% VRAM
S
MiniCPM-V 4.61.3B
OTHER·256K CTX· CHAT· VISION
105
TOK/S · 32% VRAM
S
DeepSeek Coder 1.3B1.35B
DEEPSEEK·16K CTX· CODING
101
TOK/S · 33% VRAM
S
Phi-1 1.3B1.42B
PHI·2K CTX· CODING
96
TOK/S · 34% VRAM
S
Phi-1.5 1.3B1.42B
PHI·2K CTX· CHAT· CODING
96
TOK/S · 34% VRAM
S
Qwen2.5-Coder-1.5B1.5B
QWEN·32K CTX· CHAT· TOOL_USE· CODING
91
TOK/S · 35% VRAM
▸ NEXT STEP

Get personalized recommendations.

See ranked models with benchmark scores, run commands, and precise speed estimates for your FirePro S7000.

WHAT THIS CARD IS WORTH

FirePro S7000 holds 97 of the models in our catalogue and is, in practice, a Q4_K_M card — the largest it takes is InternLM2 5B at Q4_K_M.

TOKENS/SEC PER $100
8B at Q4_K_M, so cards compare like for like
VRAM PER $100
what memory costs on this card
THE BIGGEST IT TAKES
InternLM2 5B4.5B · Q4_K_M3.6 GB25 tok/sEST
Ministral 3 3B Reasoning4.25B · Q4_K_M3.3 GB26 tok/sEST
Qwen3 4B4B · Q4_K_M3.3 GB27 tok/sEST
Qwen3-4B Instruct 25074B · Q4_K_M3.3 GB27 tok/sEST
Qwen3-Embedding 4B4B · Q4_K_M3.3 GB27 tok/sEST
Nemotron 3 Nano 4B3.97B · Q4_K_M3.4 GB27 tok/sEST
Ministral 3 3B3.85B · Q5_K_M3.5 GB25 tok/sEST
phi-3-mini-4k 3.8B3.8B · Q4_K_M3.4 GB28 tok/sEST
AGAINST CARDS OF SIMILAR MEMORY
RTX 3070 Ti8 GB$49912.6 tok/s per $100
RTX 3060 Ti GDDR6X8 GB$39915.8 tok/s per $100
RTX 3070 Ti 8 GB GA1028 GB$59910.5 tok/s per $100
Arc A7508 GB$19910.1 tok/s per $100
FirePro S70004 GB

Neighbours in memory rather than in price: memory decides whether a card can do the job at all, so two cards of the same size at different prices is the comparison you are making. Ordered by memory, then bandwidth — not by the value column, which is worked out from bandwidth and price alone and therefore rewards a cheap card whatever its software stack does to that bandwidth in practice. Read it as one input, not as a ranking.

▸ DEVICE UNDER TEST

AMD FirePro S7000 — 4 GB VRAM.

FIREPRO S7000 SPEC
BRAND
AMD
VRAM
4 GB GDDR5
BANDWIDTH
153.6 GB/s
FP16 COMPUTE
2.4 TFLOPS
FP32 COMPUTE
2.4 TFLOPS
STREAM PROCESSORS
1,280
TDP
150 W
ARCHITECTURE
GCN 1.0
▸ AI CAPABILITY
97/ 449 models @ Q4

With 4 GB VRAM and 153.6 GB/s bandwidth, this GPU handles models up to 4.25B parameters.

Speed ≈ bandwidth / model_size × efficiency. A 7B model at Q4 runs at ~18 tok/s.

§ 01TOP MODELS FOR FIREPRO S7000
97 FIT · SHOWING 20
MODELSIZEVRAM Q4TOK/SAVG
Ministral 3 3B Reasoning4.25B3.1 GB32
Qwen 1.5 4B4B2.9 GB3412.6
Qwen3 4B4B2.9 GB3440.7
Qwen3-4B Instruct 25074B2.9 GB3437.2
Qwen3-Embedding 4B4B2.9 GB34
Nemotron 3 Nano 4B3.97B2.9 GB3432.0
Ministral 3 3B3.85B2.8 GB3521.4
Phi-3.5 Mini 3.8B3.82B2.8 GB3646.6
phi-3-mini-4k 3.8B3.8B2.8 GB3630.5
Phi-4-mini 3.8B3.8B2.8 GB3649.0
Cogito 3B3.61B2.7 GB3822.1
Falcon3-3B3.23B2.5 GB4225.7
granite-4.0-h-micro 3.2B3.2B2.4 GB4318.4
Llama-3.2-3B3.2B2.4 GB4317.9
Falcon-H1 3B3.15B2.4 GB4349.5
Qwen 2.5 3B3.1B2.4 GB4437.2
SmolLM3-3B3.1B2.4 GB4430.5
Ministral 3B3B2.3 GB4629.6
StarCoder2 3B3B2.3 GB469.5
Granite 4.1 3B3B2.3 GB4616.6