Nemotron 3.5 Bolt 74B-A9B vs Qwen3.6 35B-A3B
Nemotron 3.5 Bolt 74B-A9B scores higher on the intelligence index, 21 against 19. Both take the same machine to start: the cheapest here that runs either is the Framework Desktop, 32GB, at $1,269. On it, Qwen3.6 35B-A3B is about 1.2× quicker: 56 tok/s against 48, both estimated from memory bandwidth. At 500k tokens a day the Framework Desktop, 32GB pays for itself in 17 years running Nemotron 3.5 Bolt 74B-A9B, against 83 years running Qwen3.6 35B-A3B.
| Nemotron 3.5 Bolt 74B-A9B | Qwen3.6 35B-A3B | |
|---|---|---|
| Intelligence index | 21 | 19 |
| Class | Haiku-class | Haiku-class |
| Weights | 22 GB | 22 GB |
| Needs at 32k | 22 GB | 23 GB |
| Quantisation | IQ2 | Q4_K_M |
| Parameters | 74.9B (9B active) | 36B (3B active) |
| Max context | 256k | 256k |
| API price per 1M | $0.3 in / $2 outpriced as Qwen3.6 27B | $0.05 in / $0.7 out |
| Licence | OpenMDW 1.1 | Apache 2.0 |
| Machines here that run it | 28 of 37 | 28 of 37 |
| Cheapest machine that runs it | Strix Halo Framework Desktop, 32GB $1,269 | Strix Halo Framework Desktop, 32GB $1,269 |
Run Nemotron 3.5 Bolt 74B-A9B on the Framework Desktop, 32GB · or Qwen3.6 35B-A3B
Jump to: Side by side on the Framework Desktop, 32GB · How much use it takes to pay for the machine · Memory is not what separates them · The assumptions behind both columns
Side by side on the Framework Desktop, 32GB
The cheapest machine that runs either model is the same one, so this is the pair doing the same work on the same hardware: Strix Halo Framework Desktop, 32GB, at $1,269.
| Nemotron 3.5 Bolt 74B-A9B | Qwen3.6 35B-A3B | |
|---|---|---|
| Speed at 32k | 48 tok/s estimated | 56 tok/s estimated |
| Pay-back on this machine | Pays back in 17 years | Pays back in 83 years |
| API cost per month | $6.18priced as Qwen3.6 27B | $1.38 |
Run Nemotron 3.5 Bolt 74B-A9B on the Framework Desktop, 32GB · or Qwen3.6 35B-A3B
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 Framework Desktop, 32GB. Nemotron 3.5 Bolt 74B-A9B 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 | Nemotron 3.5 Bolt 74B-A9B | Qwen3.6 35B-A3B |
|---|---|---|
| 50ka few chats a day | 175 years | 831 years |
| 200klight assistant use | 44 years | 208 years |
| 1Ma moderate coding-assistant day | 8.7 years | 42 years |
| 4Mheavy coding with an agent | 2.2 years | 10 years |
| 20Magents running most of the day | 5.2 months | 2.1 years |
Run Nemotron 3.5 Bolt 74B-A9B at 20M tokens a day · or Qwen3.6 35B-A3B
Memory is not what separates them
Nemotron 3.5 Bolt 74B-A9B needs 22 GB of memory at 32k of context and Qwen3.6 35B-A3B needs 23 GB. Every machine priced here that runs one runs the other, at 32k of context. So the choice between them is what each is good at, how fast it runs and what the same work costs on an API, not what you have to buy to hold it.
Nemotron 3.5 Bolt 74B-A9B is also head to head with Qwen3.6 27B above it on the leaderboard. Two more models need much the same memory: Qwen3 30B-A3B Instruct 2507 and Qwen3-Coder 30B-A3B. Qwen3.6 35B-A3B is also head to head with Muse Glimmer 30B below it on the leaderboard. One more model needs much the same memory: Gemma 4 31B it.
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. 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 Nemotron 3.5 Bolt 74B-A9B · every machine that runs Qwen3.6 35B-A3B · every other match-up · both against the frontier · the quickest pay-back at each level of use