/inferences/phi-4/mini/bartowski-q3_k_s-ollama

hf.co/bartowski/microsoft_Phi-4-mini-instruct-GGUF:Q3_K_S

bartowski · Q3_K_S · on ollama · quantizer build

Measured by one node on hardware we do not own, and signed by each of them. What machines here actually proved — never what the card claims.

matrix built 24 Oct 2026 · every field verified against the signature of the node that produced it, and re-verifiable by you: each signed figure carries the exact bytes the node signed (signing_payload_b64) and the signature over them, and the ed25519 public key is embedded in node_did. effective_ctx is the context length a node PROVED by recall, never the length a model card advertises
context declarednot declaredwhat the runtime reported. A node attests it was told this; nobody attests it is true.
context proved by recallnot probedthe largest size at which a machine could still find tokens planted across the whole prompt — start, middle and end, all of which had to come back.
Both numbers are needed to state a gap, and one of them is missing.

Other builds of Phi 4 MINI

every other build of this member measured on the network — compare what each actually proved
hf.co/MaziyarPanahi/Phi-4-mini-instruct-GGUF:Q3_K_Lmaziyarpanahi · Q3_K_L · ollama
1,31,072 ctx proved3 nodes
hf.co/unsloth/Phi-4-mini-instruct-GGUF:Q4_K_Munsloth · Q4_K_M · ollama
1,31,072 ctx proved2 nodes
hf.co/unsloth/Phi-4-mini-instruct-GGUF:Q5_K_Munsloth · Q5_K_M · ollama
1,31,072 ctx proved1 node
hf.co/unsloth/Phi-4-mini-instruct-GGUF:Q3_K_Munsloth · Q3_K_M · ollama
1,31,072 ctx proved1 node
hf.co/unsloth/Phi-4-mini-instruct-GGUF:Q8_0unsloth · Q8_0 · ollama
1,31,072 ctx proved1 node
hf.co/unsloth/Phi-4-mini-instruct-GGUF:BF16unsloth · BF16 · ollama
1,31,072 ctx proved2 nodes
hf.co/bartowski/microsoft_Phi-4-mini-instruct-GGUF:Q3_K_Lbartowski · Q3_K_L · ollama
8,192 ctx proved1 node
hf.co/MaziyarPanahi/Phi-4-mini-instruct-GGUF:Q3_K_Smaziyarpanahi · Q3_K_S · ollama
4,096 ctx proved3 nodes
hf.co/bartowski/microsoft_Phi-4-mini-instruct-GGUF:Q4_K_Mbartowski · Q4_K_M · ollama
4,096 ctx proved1 node
hf.co/bartowski/microsoft_Phi-4-mini-instruct-GGUF:IQ2_Mbartowski · IQ2_M · ollama
4,096 ctx proved2 nodes
hf.co/bartowski/microsoft_Phi-4-mini-instruct-GGUF:IQ4_XSbartowski · IQ4_XS · ollama
4,096 ctx proved1 node
hf.co/bartowski/microsoft_Phi-4-mini-instruct-GGUF:Q4_0bartowski · Q4_0 · ollama
4,096 ctx proved1 node
hf.co/bartowski/microsoft_Phi-4-mini-instruct-GGUF:IQ3_XXSbartowski · IQ3_XXS · ollama
4,096 ctx proved1 node
microsoft/Phi-4-mini-instructmicrosoft · BF16 · vllm
4,096 ctx proved1 node
hf.co/MaziyarPanahi/Phi-4-mini-instruct-GGUF:Q2_Kmaziyarpanahi · Q2_K · ollama
ctx not probed1 node
hf.co/MaziyarPanahi/Phi-4-mini-instruct-GGUF:Q4_K_Mmaziyarpanahi · Q4_K_M · ollama
ctx not probed1 node
hf.co/MaziyarPanahi/Phi-4-mini-instruct-GGUF:Q4_K_Smaziyarpanahi · Q4_K_S · ollama
ctx not probed1 node
hf.co/unsloth/Phi-4-mini-instruct-GGUF:Q2_Kunsloth · Q2_K · ollama
ctx not probed1 node
hf.co/MaziyarPanahi/Phi-4-mini-instruct-GGUF:Q8_0maziyarpanahi · Q8_0 · ollama
ctx not probed1 node
hf.co/Mungert/Phi-4-mini-instruct.gguf:IQ2_XSmungert · IQ2_XS · ollama
ctx not probed1 node
signed by this network

What machines here proved

peak throughput ever signed28.18 tok/s · did:epn:002408…afbb8a
machines that measured it1
samples behind those numbers1
regions it was proved in1
Every line traces to a node signature.
from the public record · unsigned

What the runtime says about it

No runtime metadata was attested for this model.The only unsigned block on this page. We attest we read these from the runtime — not that they are true. Nobody signs a model card.

Proven across regions

The machine that measured it

nothing is averaged — the machine is what you are choosing
did:epn:002408…afbb8aMuzaffarpur · 842001 · 16 Oct 2026
—ctx proved
28.18tok/s now1 samples · 2s
no capability probed