FitMyLLM
▸ AMD· GCN 2.0

AMD FirePro S9150

Running LLMs on the FirePro S9150 — the long read: which models fit at which quantisation, and the settings worth changing. · Or what a budget buys

VRAM
16 GB
MID-RANGE
BANDWIDTH
320
GB/S
MODELS Q4
262/449
58%
7B Q4 SPEED
~41
FAST
▸ MODEL COVERAGE @ Q458% OF ALL
▸ ESTIMATED SPEED· BY MODEL SIZE @ Q4

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

3B
~95
TOK/S
7B
~41
TOK/S
14B
~20
TOK/S
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
16 GB
BANDWIDTH
320 GB/s
FP16 COMPUTE
5.1 TFLOPS
TDP
235W
MEMORY
GDDR5
ARCHITECTURE
GCN 2.0
STREAM PROCESSORS
2,816
PCIE
Gen 3 x16
219
FAST MODELS · >30 TOK/S
Real-time chat speed
262
USABLE · >10 TOK/S
Comfortable for all tasks
262
TOTAL COMPATIBLE
Fit in VRAM at Q4
▸ DON’T WANT TO BUY?

Test FirePro S9150 (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· 262
S
Falcon-H1R Tiny 90M0.09B
FALCON·256K CTX· CHAT· REASONING
3160
TOK/S · 3% VRAM
›
S
SmolLM2 135M0.135B
SMOLLM·2K CTX· CHAT
2107
TOK/S · 4% VRAM
›
S
nomic-embed-text-v1.5 100M0.14B
EMBEDDING·8K CTX· CHAT
2032
TOK/S · 4% VRAM
›
S
GPT-2 124M0.14B
GPT2·1K CTX· CHAT
2032
TOK/S · 4% VRAM
›
S
SmolVLM 256M0.256B
SMOLLM·8K CTX· CHAT· VISION
1111
TOK/S · 4% VRAM
›
S
Gemma 3 270M0.27B
GEMMA·32K CTX· CHAT
1053
TOK/S · 4% VRAM
›
S
Snowflake Arctic Embed M v2.00.305B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
933
TOK/S · 4% VRAM
›
S
bge-large-en-v1.5 335M0.335B
EMBEDDING·1K CTX· CHAT
849
TOK/S · 4% VRAM
›
S
mxbai-embed-large-v10.335B
EMBEDDING·1K CTX· EMBEDDING
849
TOK/S · 4% VRAM
›
S
Snowflake Arctic Embed L0.335B
EMBEDDING·1K CTX· EMBEDDING
849
TOK/S · 4% VRAM
›
S
LFM2 350M0.35B
LFM·125K CTX· CHAT· TOOL_USE
813
TOK/S · 4% VRAM
›
S
SmolLM2 360M0.36B
SMOLLM·8K CTX· CHAT
790
TOK/S · 4% VRAM
›
S
GPT-2 Medium 345M0.38B
GPT2·1K CTX· CHAT
749
TOK/S · 5% VRAM
›
S
Qwen 2.5 0.5B0.5B
QWEN·32K CTX· CHAT
569
TOK/S · 5% VRAM
›
S
SmolVLM 500M0.5B
SMOLLM·8K CTX· CHAT· VISION
569
TOK/S · 5% VRAM
›
S
Falcon-H1 0.5B0.52B
FALCON·128K CTX· CHAT
547
TOK/S · 5% VRAM
›
S
BGE-M30.568B
EMBEDDING·8K CTX· EMBEDDING
501
TOK/S · 5% VRAM
›
S
Snowflake Arctic Embed L v2.00.568B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
501
TOK/S · 5% VRAM
›
S
Qwen3-Embedding 0.6B0.6B
EMBEDDING·32K CTX· EMBEDDING· MULTILINGUAL
474
TOK/S · 5% VRAM
›
S
Falcon-H1R Tiny 0.6B0.6B
FALCON·32K CTX· CHAT· REASONING
474
TOK/S · 5% VRAM
›
S
Falcon Perception 0.6B0.6B
FALCON·4K CTX· VISION
474
TOK/S · 5% VRAM
›
S
Qwen 1.5 0.5B0.62B
QWEN·32K CTX· CHAT
459
TOK/S · 5% VRAM
›
S
LFM2 700M0.74B
LFM·125K CTX· CHAT· TOOL_USE
384
TOK/S · 6% VRAM
›
S
Qwen3 0.6B0.75B
QWEN·32K CTX· CHAT· REASONING
379
TOK/S · 6% VRAM
›
S
GPT-2 Large 774M0.81B
GPT2·1K CTX· CHAT
351
TOK/S · 6% VRAM
›
S
Qwen 3.5 0.8B0.87B
QWEN·256K CTX· CHAT· CODING· MULTILINGUAL
327
TOK/S · 6% VRAM
›
S
InternVL3 1B0.94B
OTHER·32K CTX· CHAT· VISION
303
TOK/S · 7% VRAM
›
S
MiniCPM5 1B1.08B
MINICPM·128K CTX· CHAT· REASONING· MULTILINGUAL
263
TOK/S · 7% VRAM
›
S
TinyLlama 1.1B1.1B
LLAMA·2K CTX· CHAT
259
TOK/S · 7% VRAM
›
S
LFM2.5-1.2B-Thinking1.2B
LFM·122K CTX· CHAT· REASONING· TOOL_USE
237
TOK/S · 8% VRAM
›
S
Llama-3.2-1B1.2B
LLAMA·4K CTX· CHAT
237
TOK/S · 8% VRAM
›
S
LFM2 1.2B1.2B
LFM·125K CTX· CHAT· TOOL_USE· MULTILINGUAL
237
TOK/S · 8% VRAM
›
S
Zamba2 1.2B1.2B
OTHER·4K CTX· CHAT
237
TOK/S · 8% VRAM
›
S
EXAONE-4.0-1.2B1.3B
EXAONE·64K CTX· CHAT
219
TOK/S · 8% VRAM
›
S
OPT 1.3B1.3B
OPT·2K CTX· CHAT
219
TOK/S · 8% VRAM
›
S
MiniCPM-V 4.61.3B
OTHER·256K CTX· CHAT· VISION
219
TOK/S · 8% VRAM
›
S
DeepSeek Coder 1.3B1.35B
DEEPSEEK·16K CTX· CODING
211
TOK/S · 8% VRAM
›
S
Phi-1 1.3B1.42B
PHI·2K CTX· CODING
200
TOK/S · 8% VRAM
›
S
Phi-1.5 1.3B1.42B
PHI·2K CTX· CHAT· CODING
200
TOK/S · 8% VRAM
›
S
LFM2 8B A1B8.3BMoE
LFM·125K CTX· CHAT· TOOL_USE· MULTILINGUAL
190
TOK/S · 35% VRAM
›
▸ NEXT STEP

Get personalized recommendations.

See ranked models with benchmark scores, run commands, and precise speed estimates for your FirePro S9150.

WHAT THIS CARD IS WORTH

FirePro S9150 holds 262 of the models in our catalogue and is, in practice, a IQ4_XS card — the largest it takes is ERNIE 4.5 21B A3B at IQ4_XS.

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
ERNIE 4.5 21B A3B21.95B · IQ4_XS13.9 GB75 tok/sEST
GPT-OSS 20B21B · Q4_K_S13.8 GB63 tok/sEST
Reka Flash 321B · Q4_K_S14.2 GB16 tok/sEST
Reka Flash 3.121B · Q4_K_S14.2 GB16 tok/sEST
InternLM2 20B19.8B · Q4_K_M14.3 GB16 tok/sEST
InternLM2.5 20B19.8B · Q4_K_M14.3 GB16 tok/sEST
Ling-lite 16.8B16.8B · Q5_K_M13.6 GB74 tok/sEST
DeepSeek V2 Lite 16B16B · Q5_K_M12.9 GB74 tok/sEST
AGAINST CARDS OF SIMILAR MEMORY
M1 Max (32GB)21 GB$1,4993.0 tok/s per $100
M2 Max (32GB)21 GB$1,7992.5 tok/s per $100
M2 Pro (32GB)21 GB$1,4991.9 tok/s per $100
M4 (32GB)21 GB$1,1991.7 tok/s per $100
FirePro S915016 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 S9150 — 16 GB VRAM.

▸ FIREPRO S9150 SPEC
BRAND
AMD
VRAM
16 GB GDDR5
BANDWIDTH
320 GB/s
FP16 COMPUTE
5.1 TFLOPS
FP32 COMPUTE
5.1 TFLOPS
STREAM PROCESSORS
2,816
TDP
235 W
ARCHITECTURE
GCN 2.0
▸ AI CAPABILITY
262/ 449 models @ Q4

With 16 GB VRAM and 320 GB/s bandwidth, this GPU handles models up to 19.8B parameters.

Speed ≈ bandwidth / model_size × efficiency. A 7B model at Q4 runs at ~37 tok/s.

RENT IT — LIVE PRICES
Checking what the clouds are charging…
§ 01TOP MODELS FOR FIREPRO S9150
262 FIT · SHOWING 20
MODELSIZEVRAM Q4TOK/SAVG
InternLM2 20B19.8B12.6 GB1445.1
InternLM2.5 20B19.8B12.6 GB1450.9
Ling-lite 16.8B16.8B10.8 GB119—
DeepSeek V2 Lite 16B16B10.3 GB11938.0
StarCoder2 15B15.96B10.2 GB1826.5
DeepSeek-Coder-V2-Lite 15.7B15.7B10.1 GB11943.0
DeepSeek-VL2 Small 16B15.7B10.1 GB11943.1
StarCoder 15B15.5B10.0 GB1821.0
InternVL3 14B15.12B9.7 GB1938.1
Phi-4-reasoning-vision 15B15B9.7 GB1942.8
DeepSeek R1 Distill Qwen 14B14.8B9.5 GB1943.9
DeepCoder 14B14.8B9.5 GB1938.7
Qwen2.5-Coder-14B14.8B9.5 GB1941.3
Qwen2.5-14B14.8B9.5 GB1941.3
Qwen3 14B14.8B9.5 GB1945.7
phi-4 14B14.66B9.4 GB1933.7
Phi-4-reasoning 14B14.66B9.4 GB1933.7
Phi-4-reasoning-plus 14B14.66B9.4 GB1975.5
Ministral 3 14B14B9.0 GB2025.9
Phi-3-medium-14b14B9.0 GB2033.7