Inkling Small vs Tencent Hy3
Inkling Small and Tencent Hy3 both score 26 on the intelligence index. None of the 37 machines here still sold at a published price runs Tencent Hy3 at 32k of context. Inkling Small runs on the Mac Studio M5 Ultra, 256GB, from $10,799. At 500k tokens a day that machine pays for itself in 124 years running Inkling Small. Shorten the context to 16k and that machine holds both.
| Inkling Small | Tencent Hy3 | |
|---|---|---|
| Intelligence index | 26 | 26 |
| Class | Haiku-class | Haiku-class |
| Weights | 163 GB | 182 GB |
| Needs at 32k | 164 GB | 193 GB |
| Quantisation | UD-Q4_K_M | Q4_K_M |
| Parameters | 266B (12B active) | 295B (21B active) |
| Max context | 1M | 256k |
| API price per 1M | $0.45 in / $1.2 out | $0.0825 in / $0.33 out |
| Licence | Apache 2.0 | Apache 2.0 |
| Machines here that run it | 1 of 37 | 0 of 37 |
| Cheapest machine that runs it | Mac Studio M5 Ultra, 256GB $10,799 | none listed |
Run Inkling Small on the Mac Studio M5 Ultra, 256GB · or Tencent Hy3 at 16k of context
Jump to: Side by side on the Mac Studio M5 Ultra, 256GB at 16k of context · How much use it takes to pay for the machine · Machines that run one and not the other · The assumptions behind both columns
Side by side on the Mac Studio M5 Ultra, 256GB at 16k of context
Tencent Hy3 needs 193 GB at 32k of context. The biggest machine on this list is the Mac Studio M5 Ultra, 256GB, and it has 192 GB usable. Tencent Hy3's weights are 182 GB of that. The rest is the cache, and the cache is the part that shrinks when you ask for less context: at 16k the model needs 188 GB, which that machine holds. It already runs Inkling Small, so at 16k of context one machine here runs both. It costs $10,799. The rest of this page is what each model does with it.
| Inkling Small | Tencent Hy3 | |
|---|---|---|
| Speed at 16k | 46 tok/s estimated | 20 tok/s estimated |
| Pay-back on this machine | Pays back in 123 years | Pays back in 1,031 years |
| API cost per month | $7.56 | $1.49 |
Run Inkling Small on the Mac Studio M5 Ultra, 256GB · or Tencent Hy3
How much use it takes to pay for the machine
Everything above is at 500k tokens a day. Pay-back moves with how much you actually run, so here are both models at the five levels of use the calculator names, on the Mac Studio M5 Ultra, 256GB at 16k of context. Inkling Small pays for it sooner at every level of use, so which of them to run does not turn on how hard you work it.
| A day's use | Inkling Small | Tencent Hy3 |
|---|---|---|
| 50ka few chats a day | 1,233 years | 10,305 years |
| 200klight assistant use | 308 years | 2,576 years |
| 1Ma moderate coding-assistant day | 62 years | 515 years |
| 4Mheavy coding with an agent | 15 years | 129 years |
| 20Magents running most of the day | 3.1 years | 26 years |
Run Inkling Small at 20M tokens a day · or Tencent Hy3
Machines that run one and not the other
Inkling Small needs 164 GB of memory at 32k of context and Tencent Hy3 needs 193 GB. That puts Inkling Small on 1 of the 37 machines priced here that Tencent Hy3 does not, starting at $10,799.
| Machine | Price | Memory | Speed on Inkling Small | Pay-back |
|---|---|---|---|---|
| Mac Studio M5 Ultra, 256GB | $10,799 | 256 GB | 43 tok/s estimated | Pays back in 124 years |
Inkling Small is also head to head with Beam 1.0 501B-A23B above it on the leaderboard. One more model needs much the same memory: DeepSeek V4-Flash. Tencent Hy3 is also head to head with Ling 3.0 flash below it on the leaderboard. One more model needs much the same memory: GLM-5.3-Flash.
The assumptions behind both columns
Both columns use the same usage: 500k tokens a day at 15:1 input to output, 32k of context, $0.17 per kWh, and today's API prices held flat. The two sections that need one machine to hold both models are at 16k of context instead, which is the longest on the calculator's list where one does. Speeds marked estimated are worked out from memory bandwidth rather than measured, and pay-back scales with them. Where nobody rents an open model by the token, its API prices are the nearest hosted model's, named beside them. Machines are the 37 here with a published price that are still sold. Change any of it in the calculator.
More head to head: every machine that runs Inkling Small · what Tencent Hy3 needs · every other match-up · both against the frontier · the quickest pay-back at each level of use