Hunyuan MT 2 1.8B
222 tokens a second on Windows · GeForce RTX 3080 Ti 12 GB · 16 CPUs · 16 GB RAM. About $1.16 per million tokens here, against OpenRouter's $0.18: the API is 6.6× cheaper. Run it here for privacy or to keep work on your own machines, not to save money.
1 build of Hunyuan MT 2 1.8B, 4 signed runs on real machines: different packagers, precisions and engines, each a different file with its own numbers.
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
Same build, different machines
Q4_K_M by tencent on 4 machines: the fastest decodes 19× as fast as the slowest. One dot per machine, tokens per second.
What it fits on
The smallest build needs about 3.2 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 cardevery build
- NVIDIA GeForce GTX 16504 GB on the cardevery build
- AMD Radeon(TM) Vega 8 Mobile Graphics7 GB RAMevery build
- Windows, CPU only · 12 CPUs7 GB RAMevery build
- Apple silicon8 GB RAMevery build
- Apple M18 GB RAMevery build
- Mac, CPU only · 8 CPUs8 GB RAMevery build
- 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 · trained for 256K. 4 walks ended because the model forgot, 0 because the machine could not hold the prompt.
| machine · build | 4K | why it stopped |
|---|---|---|
| Gram ·84bf54NVIDIA GeForce RTX 3080 Ti · Q4_K_M by tencent | — the model lost the needle at 4K | |
| Gram ·dae679NVIDIA L4 · Q4_K_M by tencent | — the model lost the needle at 4K | |
| Gram ·2e5c02Intel(R) Iris(R) Xe Graphics · Q4_K_M by tencent | — the model lost the needle at 4K | |
| Gram ·9e3c79Intel(R) UHD Graphics 730 · Q4_K_M by tencent | — the model lost the needle at 4K |
Carbon per million tokens
Lowest build on each machine · 2 of 4 read off a power sensor, the rest modelled from the parts · fewer grams is better
- Gram ·dae679438 g3.33 J/tok · sensor
- Gram ·2e5c024.1 kg21.08 J/tok · modelled
- Gram ·9e3c7922 kg111.04 J/tok · modelled
- Gram ·84bf54130 kg663.33 J/tok · sensor
Carbon is location-based: signed energy × the yearly average of the grid where the machine signed its place (world grid average (Singapore not in the table) · 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
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 |
|---|---|---|---|---|---|---|
| Q4_K_M by tencenthf.co/tencent/Hy-MT2-1.8B-GGUF:Q4_K_M | ollama | vendor · tencent | 3.2 GB | — | 233.3 | 4 |
Fine-tunes, merges and abliterations of this member, each its own model with its own page. None of their numbers are counted above.