family member
Llama 3.2 logo

Llama 3.2 1B

20 tokens a second on Mac · Apple silicon · 8 CPUs · 8 GB RAM. About $2.66 per million tokens here, against OpenRouter's $0.20: the API is 13× cheaper. Run it here for privacy or to keep work on your own machines, not to save money.

9 builds of Llama 3.2 1B, 28 signed runs on real machines: different packagers, precisions and engines, each a different file with its own numbers. 1.2B parameters, as the vendor states.

20tok/s, fastest machine
$2.66per M tokens · $0.20 on an API
32Kcontext proved by recall
0.7 GBmemory, smallest build
9providers measured it
Use it in your own room Ask Evo in your gramx and it runs on a machine that proved it, at the metered $2.66 per million tokens. On your own machines it costs you nothing.Measure it on your machine A gram works through the catalog by itself and signs every result. 9 providers have measured Llama 3.2 1B; every run is a point on the providers board.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 · 9 builds

proved 32K context, by ollama Q8_0

113.3 tok/s peak, by MaziyarPanahi Q4_K_M

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 · dashed: OpenRouter

$1$10USD per million tokenstok/sMac · Apple silicon · 8 CPUs · 8 GB RAM · llama3.2:1b · 20.37 tok/s · $8.5986/MWindows · no GPU · 8 CPUs · 16 GB RAM · 3-bit by MaziyarPanahi · 17.72 tok/s · $4.0226/MWindows · no GPU · 8 CPUs · 16 GB RAM · 4-bit by bartowski · 11.62 tok/s · $6.161/MLinux · no GPU · 4 CPUs · 31 GB RAM · llama3.2:1b · 11.45 tok/s · $2.6566/MWindows · AMD Radeon(TM) Vega 8 Mobile Graphics 1 GB · 8 CPUs · 7 GB RAM · llama3.2:1b · 5.62 tok/s · $31.1559/MMac · no GPU · 4 CPUs · 8 GB RAM · 4-bit by bartowski · 4.73 tok/s · $6.5625/MMac · Apple silicon · 4 CPUs · 8 GB RAM · llama3.2:1b · 3.37 tok/s · $39.9708/MMac · no GPU · 4 CPUs · 8 GB RAM · 1-bit by MaziyarPanahi · 3.26 tok/s · $9.4216/MMac · no GPU · 4 CPUs · 8 GB RAM · 3-bit by MaziyarPanahi · 1.48 tok/s · $20.8155/M

Same build, different machines

Q8_0 by ollama on 10 machines: the fastest decodes 7.6× as fast as the slowest. One dot per machine, tokens per second.

  1. Apple M1×229 tok/s
  2. Intel(R) Iris(R) Xe Graphics×113 tok/s
  3. Linux, CPU only · 4 CPUs×212 tok/s
  4. 2× Tesla T4×111 tok/s
  5. Intel(R) UHD Graphics 730×19.7 tok/s
  6. AMD Radeon(TM) Vega 8 Mobile Graphics×18.3 tok/s
  7. Apple M4×16.2 tok/s
  8. apple×13.8 tok/s

What it fits on

The smallest build needs about 0.7 GB; 13 of 13 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.

  1. NVIDIA GeForce RTX 3050 Laptop GPU4 GB on the cardevery build
  2. NVIDIA GeForce GTX 16504 GB on the cardevery build
  3. AMD Radeon(TM) Vega 8 Mobile Graphics7 GB RAMevery build
  4. Windows, CPU only · 12 CPUs7 GB RAMevery build
  5. apple8 GB RAMevery build
  6. Apple M18 GB RAMevery build
  7. NVIDIA GeForce RTX 3080 Ti12 GB on the cardevery build
  8. Tesla T415 GB on the cardevery build
  9. Intel(R) UHD Graphics 73016 GB RAMevery build
  10. Intel(R) Iris(R) Xe Graphics16 GB RAMevery build
  11. Apple M416 GB RAMevery build
  12. 2× Tesla T429 GB on the cardevery build
  13. Linux, CPU only · 4 CPUs31 GB RAMevery 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 · trained for 128K. 3 walks ended because the model forgot, 1 because the machine could not hold the prompt.

machine · build4K8K16K32K64K128Kwhy it stopped
Gram ·0548a8Apple M1 · Q8_0 by ollama32K
nitin singhApple M4 · Q4_K_M by bartowski8K
nitin singhApple M4 · Q8_0 by ollama8K
nitin singhApple M4 · IQ3_XS by MaziyarPanahi4K
Amit KhetanIntel(R) UHD Graphics 730 · Q4_K_M by bartowski4K
Gram ·d061c9AMD Radeon(TM) Vega 8 Mobile Graphics · Q8_0 by ollama4K too slow to answer here
nitin singhApple M4 · IQ1_S by MaziyarPanahi—
Amit KhetanIntel(R) UHD Graphics 730 · IQ3_XS by MaziyarPanahi—
Keshav AgarwalIntel(R) Iris(R) Xe Graphics · Q8_0 by ollama—
Gram ·e3654fLinux, CPU only · 4 CPUs · Q8_0 by ollama— the model lost the needle at 4K
Gram ·acbe75Linux, CPU only · 4 CPUs · Q8_0 by ollama— the model lost the needle at 4K
Amit KhetanIntel(R) UHD Graphics 730 · IQ1_S by MaziyarPanahi—
Amit KhetanIntel(R) UHD Graphics 730 · Q8_0 by ollama—
Amit KhetanIntel(R) UHD Graphics 730 · Q4_K_M by MaziyarPanahi—

The 14 longest of 19 walks; every one is on its build's page.

Carbon per million tokens

Lowest build on each machine · 0 of 5 read off a power sensor, the rest modelled from the parts · fewer grams is better

  1. Gram ·e3654f340 g3.19 J/tok · modelled
  2. Gram ·e61d3b424 g3.23 J/tok · modelled
  3. Gram ·acbe75477 g3.63 J/tok · modelled
  4. Amit Khetan1.1 kg5.58 J/tok · modelled
  5. Gram ·d061c93.4 kg17.52 J/tok · modelled

Carbon is location-based: signed energy × the yearly average of the grid where the machine signed its place (United States grid average · world grid average (Singapore not in the table) · world grid average (place not signed) · India grid average; Ember Global Electricity Review 2025 (2024 generation), location-based). Facility overhead is not included. A machine on its own solar is not credited until it can prove it.

Measured by

The people whose machines signed these numbers

proved by at least one build: Tool calling · Tool loop finished · Faithful to tool results · Structured output · Sustained output · Writes code · Synthesises sources · Follows instructions

Every build

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

buildenginewhose workmemorycontext provedbest tok/smachines
Q8_0 by ollamallama3.2:1bollamaollama2.9 GB32K49.910
Q4_K_M by bartowskihf.co/bartowski/Llama-3.2-1B-Instruct-GGUF:Q4_K_Mollamaquantizer · bartowski1.1 GB16K103.94
IQ3_XS by MaziyarPanahihf.co/MaziyarPanahi/Llama-3.2-1B-Instruct-GGUF:IQ3_XSollamaquantizer · MaziyarPanahi—4K107.53
Q8_0 by bartowskihf.co/bartowski/Llama-3.2-1B-Instruct-GGUF:Q8_0ollamaquantizer · bartowski1.7 GB4K8.71
Q4_K_M by MaziyarPanahihf.co/MaziyarPanahi/Llama-3.2-1B-Instruct-GGUF:Q4_K_Mollamaquantizer · MaziyarPanahi——113.32
IQ1_S by MaziyarPanahihf.co/MaziyarPanahi/Llama-3.2-1B-Instruct-GGUF:IQ1_Sollamaquantizer · MaziyarPanahi0.7 GB—105.04
Q3_K_L by bartowskihf.co/bartowski/Llama-3.2-1B-Instruct-GGUF:Q3_K_Lollamaquantizer · bartowski1.0 GB—10.41
F16 by bartowskihf.co/bartowski/Llama-3.2-1B-Instruct-GGUF:F16ollamaquantizer · bartowski3.0 GB—8.02
IQ3_M by bartowskihf.co/bartowski/Llama-3.2-1B-Instruct-GGUF:IQ3_Mollamaquantizer · bartowski1.0 GB—5.31
derivatives — somebody else's change to these weights

Fine-tunes, merges and abliterations of this member, each its own model with its own page. None of their numbers are counted above.

Llama-3.2-1B-Instruct-Q4_K_M-GGUF1 build measuredLlama-3.2-1B-Instruct-bf16not measuredLlama-3.2-1B-Instruct-LinkLlama-Cap50not measuredLlama-3.2-1B-Instructnot measured