/inferences/llama-3-2/1b/ollama-q8_0-ollama

llama3.2:1b

ollama · Q8_0 · on ollama

Measured by 10 nodes on hardware we do not own, and signed by each of them. What machines here actually proved — never what the card claims.

matrix built 26 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 declared1,31,072what the runtime reported. A node attests it was told this; nobody attests it is true.
context proved by recall32,768the 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.
Advertises 4× the context any machine here could make it use.Not an accusation — it is what happened here, and it is signed.
recalled 32k⌐ declared 128k · unsignedadvertises 4× what it proved
every machine that timed it
29.1 tok/s · unnamed machine20.4 tok/s · unnamed machine13.1 tok/s · unnamed machine11.4 tok/s · x86 · CPU · North Charleston11.0 tok/s · x86 · CPU · Kane10.7 tok/s · unnamed machine9.7 tok/s · unnamed machine7.7 tok/s · unnamed machine3.4 tok/s · Apple silicon · Bhubaneswar0.8 tok/s · unnamed machine
0.8 → 29.1 tok/s across 10 machines · ×36.1 spreadunnamed machinex86 · CPUApple silicon

This model calls tools correctly and does not stop. A node watched it call the probe tool, then call it again, until the turn budget ran out. Chosen on “supports tools” alone, it will call a tool in production and never come back.

Languages, proved

BengaliEnglishHindiKannadaTelugu

Each question was put entirely in its own language and script, and the answer had to be right. A model with no coverage cannot parse the question, so answering in English is a failure here, not a pass. 5 proved of 13 asked — a language absent from the probe set was never asked, and is not a claim either way.

Capabilities, measured

Audioprobed, and failed
Given audio and asked to transcribe it.
Reasons about codeprobed, and failed
Asked to predict exactly what a short program prints. This is the half of coding that matters for agent work and that writing valid syntax does not demonstrate: following state through control flow.
Writes codepassed
Asked for a specific function and checked with the Go compiler’s own parser. Nothing is executed — running model-written code to score a model would be a security hole opened for a metric — but syntactic validity is a fact rather than an opinion.
Follows instructionspassed
Given a plain constraint on its output format and checked for exact obedience. This separates an instruct-tuned model from a base one — a base model answers fluently and ignores every instruction, which then looks like a dozen unrelated weaknesses instead of one.
Sustained outputpassed
Asked to produce a long structured answer and checked for length. A model that stops after a couple of hundred tokens cannot write a report, however capable it is otherwise.
Structured outputpassed
The ENGINE was asked to constrain decoding to a schema, and the answer conformed. This is stronger than politely asking for JSON and hoping: the grammar makes invalid output unreachable, which is what an operator actually builds on.
Extended thinkingprobed, and failed
The model emitted reasoning on the engine’s dedicated reasoning channel before answering. Read from that channel rather than by scanning the answer for tags, which is why models that reason are no longer reported as models that do not.
Synthesises sourcespassed
Given three documents with one fact split across them, and required to combine rather than quote. No single document contains the answer, so a model that retrieves without reasoning cannot pass.
Tool callingpassed
The node asked the model to call a function and checked that it called the RIGHT one with a well-formed argument — not merely that it emitted a tool name.
Faithful to tool resultspassed
The model was handed a list through a tool and had to report every item back. Calling a tool and understanding its answer are different abilities: a model can produce a perfect call, receive nine items, and confidently report three. This is the flag that separates them.
Tool loop finishedpassed
Whether the conversation ENDED after the tool call, or the model kept calling until the turn budget ran out. A model that passes tool calling and fails this will call a tool in production and never come back.
Chained tool callsprobed, and failed
The model had to call one tool, read its answer, and call a second with a value derived from it. One call is not agentic work — real specialist jobs are sequences, and a model that never takes the second step looks, from outside, exactly like a specialist silently doing nothing.
Embeddingsprobed, and failed
The engine was asked for an embedding of a short text and returned a real, non-zero vector.
Visionprobed, and failed
A four-digit code was rendered into an image and the model had to read it back. A model that cannot see cannot guess four digits, and a model that can has no reason to fail — so this passes and fails for the right reason, which describing a picture never did.
Speech outprobed, and failed
The model itself turned text into audio — identified from the container bytes, not from what the engine labelled them, so an error page announced as audio cannot pass. Distinct from this gram running a separate text-to-speech engine: an operator choosing between them is choosing between two architectures.

Other builds of Llama 3.2 1B

every other build of this member measured on the network — compare what each actually proved
hf.co/bartowski/Llama-3.2-1B-Instruct-GGUF:Q4_K_Mbartowski · Q4_K_M · ollama
16,384 ctx proved4 nodes
hf.co/MaziyarPanahi/Llama-3.2-1B-Instruct-GGUF:IQ3_XSMaziyarPanahi · IQ3_XS · ollama
4,096 ctx proved3 nodes
hf.co/bartowski/Llama-3.2-1B-Instruct-GGUF:Q8_0bartowski · Q8_0 · ollama
4,096 ctx proved1 node
hf.co/MaziyarPanahi/Llama-3.2-1B-Instruct-GGUF:Q4_K_MMaziyarPanahi · Q4_K_M · ollama
ctx not probed2 nodes
hf.co/MaziyarPanahi/Llama-3.2-1B-Instruct-GGUF:IQ1_SMaziyarPanahi · IQ1_S · ollama
ctx not probed4 nodes
hf.co/bartowski/Llama-3.2-1B-Instruct-GGUF:Q3_K_Lbartowski · Q3_K_L · ollama
ctx not probed1 node
hf.co/bartowski/Llama-3.2-1B-Instruct-GGUF:F16bartowski · F16 · ollama
ctx not probed2 nodes
hf.co/bartowski/Llama-3.2-1B-Instruct-GGUF:IQ3_Mbartowski · IQ3_M · ollama
ctx not probed1 node
signed by this network

What machines here proved

context proved by recall32,768
peak throughput ever signed49.89 tok/s · did:epn:002408…afbb8a
resident while loaded2.4 GiB (2.4 on card) · did:epn:002408…0548a8
machines that measured it10
samples behind those numbers10
regions it was proved in9
Every line traces to a node signature.
from the public record · unsigned

What the runtime says about it

context declared1,31,072
parameters1.2B
parameter count1,23,58,14,432
quantisationQ8_0
familyllama
formatgguf
ollama /api/show; the node attests that the runtime reported these, not that they are true. Attested by did:epn:002408…d68672.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

Hyderabad, Telangana500062 · 1 node
Hyderabad, Telangana500001 · 1 node
Vikarabad, Telangana501202 · 1 node
North Charleston, South Carolina29415 · 1 node
Kane, Iowa51503 · 1 node
Taipei, Taipei110 · 1 node
Bengaluru, Karnataka562130 · 2 nodes
Sundargarh, Odisha770033 · 1 node
Muzaffarpur, Bihar842001 · 1 node

Every machine that measured it

nothing is averaged — the machine is what you are choosing
did:epn:002408…0c5265Hyderabad · 500062 · 8 Oct 2026
—ctx proved
29.06tok/s now111 samples · 277s
no capability probed
did:epn:002408…0548a8Hyderabad · 500001 · 0.32.5 · 8 Oct 2026
32,768ctx provedprobe Is8dkyzvxI…
20.37tok/s nowpeak 28.21 · 8 Oct 2026168 samples · 628s2.4 GiB resident · 2.4 GiB on card · 1.2 GiB on disk
✗ Reasons about code✓ Writes code✗ Follows instructions✓ Sustained output✓ Structured output✗ Extended thinking✓ Synthesises sources✓ Tool calling✗ Faithful to tool results✓ Tool loop finished✗ Vision✗ Audio✓ 2 languages
did:epn:002408…2e5c02Vikarabad · 501202 · 0.32.5 · 24 Aug 2026
0ctx provedno recall at 4,096
13.09tok/s now4 samples · 11s
✗ Reasons about code✓ Writes code✓ Follows instructions✓ Sustained output✓ Structured output✗ Extended thinking✗ Synthesises sources✓ Tool calling✗ Faithful to tool results✓ Tool loop finished✗ Vision✗ Audio✓ 1 languages
did:epn:002408…e3654fNorth Charleston · 29415 · 0.40.11 · on host · 25 Oct 2026x86_64 · none · 22.5 GiB pooled
0ctx provedthe model lost the needle at 8K
11.45tok/s now62 samples · 317s1.4 GiB resident · 1.2 GiB on disk
✗ Speech out✗ Reasons about code✓ Writes code✗ Embeddings✓ Follows instructions✓ Sustained output✓ Structured output✗ Extended thinking✗ Synthesises sources✓ Tool calling✓ Faithful to tool results✓ Tool loop finished✗ Chained tool calls✗ Vision✗ Audio✓ 3 languages
did:epn:002408…e61d3bKane · 51503 · 0.40.11 · on host · 25 Oct 2026x86_64 · none · 23.4 GiB pooled
0ctx provedthe model lost the needle at 8K
10.95tok/s now59 samples · 417s1.4 GiB resident · 1.2 GiB on disk
✗ Speech out✗ Reasons about code✓ Writes code✗ Embeddings✓ Follows instructions✓ Sustained output✓ Structured output✗ Extended thinking✗ Synthesises sources✓ Tool calling✓ Faithful to tool results✓ Tool loop finished✗ Chained tool calls✗ Vision✗ Audio✓ 2 languages
did:epn:002408…acbe75Taipei · 110 · 0.40.11 · on host · 24 Oct 2026
0ctx provedthe model lost the needle at 8K
10.70tok/s nowpeak 11.28 · 24 Oct 202665 samples · 268s1.4 GiB resident · 1.2 GiB on disk
✗ Speech out✗ Reasons about code✓ Writes code✗ Embeddings✓ Follows instructions✓ Sustained output✓ Structured output✗ Extended thinking✗ Synthesises sources✓ Tool calling✓ Faithful to tool results✓ Tool loop finished✗ Chained tool calls✗ Vision✗ Audio✓ 3 languages
did:epn:002408…9e3c79Bengaluru · 562130 · ollama · 20 Aug 2026
0ctx provedno recall at 4,096
9.73tok/s now94 samples · 629s
✗ Reasons about code✓ Writes code✗ Follows instructions✓ Sustained output✓ Structured output✗ Extended thinking✓ Synthesises sources✓ Tool calling✓ Faithful to tool results✓ Tool loop finished✗ Chained tool calls✗ Vision✗ Audio✓ 1 languages
did:epn:002408…d061c9Bengaluru · 562130 · 0.34.0 · 15 Sept 2026
4,096ctx provedprobe UeUMFbNCnm…
7.70tok/s nowpeak 8.34 · 15 Sept 202615 samples · 99s1.4 GiB resident · 1.2 GiB on disk
✗ Reasons about code✓ Writes code✓ Follows instructions✓ Sustained output✓ Structured output✗ Extended thinking✗ Synthesises sources✓ Tool calling✓ Faithful to tool results✓ Tool loop finished✗ Chained tool calls✗ Vision✗ Audio✓ 2 languages
did:epn:002408…d68672Sundargarh · 770033 · 0.30.11 · on host · 24 Oct 2026Intel Core i5-5350U @ 1.80GHz · 2 cores · 8.0 GiB RAM
0ctx provedthe model lost the needle at 4K
3.37tok/s nowpeak 3.85 · 26 Sept 2026217 samples · 6,548s1.4 GiB resident · 1.2 GiB on disk
✗ Reasons about code✓ Writes code✗ Embeddings✗ Follows instructions✓ Structured output✗ Extended thinking✓ Tool calling✗ Faithful to tool results✓ Tool loop finished✗ Vision✗ Audio✓ 2 languages
did:epn:002408…afbb8aMuzaffarpur · 842001 · 0.32.14 · 20 Aug 2026
8,192ctx provedprobe yW7BFtGj9D…
0.80tok/s nowpeak 6.20 · 15 Oct 2026167 samples · 13,482s
✓ Tool calling✓ Tool loop finished✓ Structured output✗ Extended thinking✗ Vision✗ Audio

Context ladders attempted: did:epn:002408…0548a8 131072,65536,32768,16384,8192,4096 · did:epn:002408…2e5c02 4096 · did:epn:002408…e3654f 4096,8192 · did:epn:002408…e61d3b 4096,8192 · did:epn:002408…acbe75 4096,8192 · did:epn:002408…9e3c79 4096 · did:epn:002408…d061c9 4096 · did:epn:002408…d68672 4096 · did:epn:002408…afbb8a 131072,65536,32768,16384,8192,4096

Run this where your data already is

Tell us the workload, whose machines it should run on, and which jurisdiction it answers to.

Run a gramPut a machine on the network in this region; it signs its capacity into the room.Open a gramxStart your own room — an org, a community, a brand.Be invitedAccept a link from an admin to cross into the private room.