family member

12b~yuxinlu1

4 builds of 12b~yuxinlu1, 4 signed runs on real machines: different packagers, precisions and engines, each a different file with its own numbers.

26tok/s, fastest machine
128Kcontext proved by recall
7.6 GBmemory, smallest build
1provider 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. 1 provider has measured 12b~yuxinlu1 so far.Use it in your own room Ask Evo in your gramx and it runs on a machine that proved it. 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 · 4 builds

trained 256K · proved 128K context, by yuxinlu1 Q4_K_M

25.8 tok/s peak, by yuxinlu1 Q4_K_M

What it fits on

The smallest build needs about 7.6 GB; 13 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.

  1. Linux, CPU only · 2 CPUs1 GB RAMnone
  2. NVIDIA GeForce RTX 3050 Laptop GPU4 GB on the cardnone
  3. NVIDIA GeForce GTX 16504 GB on the cardnone
  4. AMD Radeon(TM) Vega 8 Mobile Graphics7 GB RAMnone
  5. Windows, CPU only · 12 CPUs7 GB RAMnone
  6. Apple silicon8 GB RAM1 of 4
  7. Apple M18 GB RAM1 of 4
  8. Mac, CPU only · 8 CPUs8 GB RAM1 of 4
  9. NVIDIA GeForce RTX 3080 Ti12 GB on the card2 of 4
  10. Tesla T415 GB on the card2 of 4
  11. Intel(R) UHD Graphics 73016 GB RAM2 of 4
  12. Intel(R) Iris(R) Xe Graphics16 GB RAM2 of 4
  13. Apple M416 GB RAM2 of 4
  14. Apple M116 GB RAM2 of 4
  15. NVIDIA L422 GB on the cardevery build
  16. 2× Tesla T429 GB on the cardevery build
  17. Linux, CPU only · 4 CPUs31 GB RAMevery build
  18. 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 · trained for 256K. 1 walk ended because the model forgot, 0 because the machine could not hold the prompt.

machine · build4K8K16K32K64K128Kwhy it stopped
Gram ·a5be462× Tesla T4 · Q6_K by yuxinlu1128K recalled at every size asked
Gram ·a5be462× Tesla T4 · Q8_0 by yuxinlu1128K recalled at every size asked
Gram ·a5be462× Tesla T4 · Q3_K_M by yuxinlu1128K recalled at every size asked
Gram ·a5be462× Tesla T4 · Q4_K_M by yuxinlu14K the model lost the needle at 8K

Measured by

The people whose machines signed these numbers

proved by at least one build: Tool calling · Tool loop finished · Faithful to tool results · Chained tool calls · Structured output · Sustained output · Extended thinking · Writes code · Reasons about 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
Q4_K_M by yuxinlu1hf.co/yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF:Q4_K_Mollamayuxinlu18.9 GB128K25.81
Q6_K by yuxinlu1hf.co/yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF:Q6_Kollamayuxinlu118.7 GB128K20.01
Q8_0 by yuxinlu1hf.co/yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF:Q8_0ollamayuxinlu118.2 GB128K18.01
Q3_K_M by yuxinlu1hf.co/yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF:Q3_K_Mollamayuxinlu17.6 GB128K17.01