family member

Phi 4 MINI-REASONING

189 tokens a second on Windows · NVIDIA 12 GB · 16 CPUs · 16 GB RAM. No hosted API sells it, so this network is the place to run it.

10 builds of Phi 4 MINI-REASONING, 35 signed runs on real machines: different packagers, precisions and engines, each a different file with its own numbers.

189tok/s, fastest machine

$1.36per M tokens · no API sells it

4Kcontext proved by recall

1.3 GBmemory, smallest build

9providers measured it

Run it on your gram Install a gram with one command. It works through the catalog on your own machine and signs every result. 9 providers have measured Phi 4 MINI-REASONING so far.Use it in your own room Ask Evo in your gramx and it runs on a machine that proved it, at the metered $1.36 per million tokens. On your own machines it costs you nothing.Compare every build Put it beside every other model in one table: speed, price, context and the API price, one row each.

best on ollama · 10 builds

proved 4K context, by unsloth IQ3_XXS

189.4 tok/s peak, by unsloth IQ3_XXS

Speed against price, every machine

Up and left is faster and cheaper · the line joins machines no other beats on both · price on a log scale

Same build, different machines

IQ1_S by unsloth on 6 machines: the fastest decodes 6.9× as fast as the slowest. One dot per machine, tokens per second.

Apple M4×128 tok/s
NVIDIA GeForce RTX 3050 Laptop GPU×120 tok/s
Mac, CPU only · 8 CPUs×116 tok/s
Intel(R) UHD Graphics 730×15.6 tok/s
Apple M1×15.6 tok/s
Intel(R) Iris(R) Xe Graphics×14.0 tok/s

What it fits on

The smallest build needs about 1.3 GB; 17 of 18 kinds of machine on the network can hold it. Bars: each kind's memory (its card's, else RAM) · ticks: every build, at the size it held plus a fifth for the prompt.

Linux, CPU only · 2 CPUs1 GB RAMnone
NVIDIA GeForce RTX 3050 Laptop GPU4 GB on the card2 of 4
NVIDIA GeForce GTX 16504 GB on the card2 of 4
AMD Radeon(TM) Vega 8 Mobile Graphics7 GB RAM3 of 4
Windows, CPU only · 12 CPUs7 GB RAM3 of 4
Apple silicon8 GB RAM3 of 4
Apple M18 GB RAM3 of 4
Mac, CPU only · 8 CPUs8 GB RAM3 of 4
NVIDIA GeForce RTX 3080 Ti12 GB on the cardevery build
Tesla T415 GB on the cardevery build
Intel(R) UHD Graphics 73016 GB RAMevery build
Intel(R) Iris(R) Xe Graphics16 GB RAMevery build
Apple M416 GB RAMevery build
Apple M116 GB RAMevery build
NVIDIA L422 GB on the cardevery build
2× Tesla T429 GB on the cardevery build
Linux, CPU only · 4 CPUs31 GB RAMevery build
NVIDIA RTX PRO 6000 Blackwell Server Edition95 GB on the cardevery build

How far back it remembers, machine by machine

Filled: the needle came back at that prompt size · outlined: asked and lost · blank: not asked. 0 walks ended because the model forgot, 0 because the machine could not hold the prompt.

machine · build	4K	8K	16K	32K	64K	128K	why it stopped
Gram ·84bf54 NVIDIA GeForce RTX 3080 Ti · IQ3_XXS by unsloth							4K an older probe: where it stopped was not recorded
Gram ·afbb8a Apple M4 · IQ3_XXS by unsloth							4K an older probe: where it stopped was not recorded
Gram ·afbb8a Apple M4 · Q8_0 by lmstudio-community							4K an older probe: where it stopped was not recorded
Gram ·84bf54 NVIDIA GeForce RTX 3080 Ti · Q3_K_L by lmstudio-community							— an older probe: where it stopped was not recorded
Gram ·53b229 NVIDIA GeForce RTX 3050 Laptop GPU · IQ3_XXS by unsloth							— an older probe: where it stopped was not recorded
Gram ·afbb8a Apple M4 · Q2_K by unsloth							— an older probe: where it stopped was not recorded
Gram ·afbb8a Apple M4 · Q3_K_M by unsloth							— an older probe: where it stopped was not recorded
Gram ·afbb8a Apple M4 · IQ1_S by unsloth							— an older probe: where it stopped was not recorded
Gram ·53b229 NVIDIA GeForce RTX 3050 Laptop GPU · IQ1_S by unsloth							— an older probe: where it stopped was not recorded
Gram ·0c5265 Mac, CPU only · 8 CPUs · IQ1_S by unsloth							— an older probe: where it stopped was not recorded
Gram ·9e3c79 Intel(R) UHD Graphics 730 · Q2_K by unsloth							— an older probe: where it stopped was not recorded
Gram ·0548a8 Apple M1 · Q4_K_M by lmstudio-community							— an older probe: where it stopped was not recorded
Gram ·9e3c79 Intel(R) UHD Graphics 730 · Q2_K_L by unsloth							— an older probe: where it stopped was not recorded
Gram ·9e3c79 Intel(R) UHD Graphics 730 · Q4_K_M by unsloth							— an older probe: where it stopped was not recorded

The 14 longest of 30 walks; every one is on its build's page.

Measured by

The people whose machines signed these numbers

proved by at least one build: Structured output · Sustained output · Extended thinking · Reasons about code · Synthesises sources

Every build

Each its own file with its own page: the machines that ran it and their signatures.

build	engine	whose work	memory	context proved	best tok/s	machines
IQ3_XXS by unslothhf.co/unsloth/Phi-4-mini-reasoning-GGUF:IQ3_XXS	ollama	quantizer · unsloth	11.5 GB	4K	189.4	6
Q8_0 by lmstudio-communityhf.co/lmstudio-community/Phi-4-mini-reasoning-GGUF:Q8_0	ollama	lmstudio-community	5.2 GB	4K	22.2	1
IQ1_S by unslothhf.co/unsloth/Phi-4-mini-reasoning-GGUF:IQ1_S	ollama	quantizer · unsloth	1.3 GB	—	188.9	9
Q3_K_L by lmstudio-communityhf.co/lmstudio-community/Phi-4-mini-reasoning-GGUF:Q3_K_L	ollama	lmstudio-community	—	—	89.9	4
Q2_K by unslothhf.co/unsloth/Phi-4-mini-reasoning-GGUF:Q2_K	ollama	quantizer · unsloth	—	—	37.7	3
Q3_K_M by unslothhf.co/unsloth/Phi-4-mini-reasoning-GGUF:Q3_K_M	ollama	quantizer · unsloth	—	—	31.9	4
Q4_K_M by lmstudio-communityhf.co/lmstudio-community/Phi-4-mini-reasoning-GGUF:Q4_K_M	ollama	lmstudio-community	—	—	7.3	2
Q2_K_L by unslothhf.co/unsloth/Phi-4-mini-reasoning-GGUF:Q2_K_L	ollama	quantizer · unsloth	—	—	6.8	3
Q4_K_M by unslothhf.co/unsloth/Phi-4-mini-reasoning-GGUF:Q4_K_M	ollama	quantizer · unsloth	2.8 GB	—	6.5	2
Q6_K by lmstudio-communityhf.co/lmstudio-community/Phi-4-mini-reasoning-GGUF:Q6_K	ollama	lmstudio-community	—	—	—	1