Mistral Small 4 (119B-2603) vs Qwen3 30B-A3B Instruct 2507
Mistral Small 4 (119B-2603) scores higher on the intelligence index, 11 against 10. The cheapest machine here that runs Mistral Small 4 (119B-2603) is the Framework Desktop, 128GB, at $3,449. Qwen3 30B-A3B Instruct 2507 runs on the Framework Desktop, 32GB at $1,269, $2,180 less. On the Framework Desktop, 128GB, the cheapest machine here that runs both, Mistral Small 4 (119B-2603) is about 1.3× quicker: 32 tok/s against 25, Qwen3 30B-A3B Instruct 2507's measured and the other estimated from memory bandwidth. At 500k tokens a day the Framework Desktop, 128GB pays for itself in 114 years running Mistral Small 4 (119B-2603), against 215 years running Qwen3 30B-A3B Instruct 2507.
| Mistral Small 4 (119B-2603) | Qwen3 30B-A3B Instruct 2507 | |
|---|---|---|
| Intelligence index | 11 | 10 |
| Class | Below every hosted tier | Below every hosted tier |
| Weights | 74 GB | 19 GB |
| Needs at 32k | 75 GB | 22 GB |
| Quantisation | Q4_K_M | Q4_K_M |
| Parameters | 119.4B (6B active) | 30.5B (3.3B active) |
| Max context | 1M | 256k |
| API price per 1M | $0.15 in / $0.6 out | $0.09 in / $0.3 out |
| Licence | Apache 2.0 | Apache 2.0 |
| Machines here that run it | 12 of 37 | 28 of 37 |
| Cheapest machine that runs it | Strix Halo Framework Desktop, 128GB $3,449 | Strix Halo Framework Desktop, 32GB $1,269 |
Run Mistral Small 4 (119B-2603) on the Framework Desktop, 128GB · or Qwen3 30B-A3B Instruct 2507 on the Framework Desktop, 32GB
Jump to: Side by side on the Framework Desktop, 128GB · 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 Framework Desktop, 128GB
The table above gives each model the cheapest machine that runs it, and those are two different machines, so nothing in it is a like-for-like race. The Strix Halo Framework Desktop, 128GB is the cheapest machine here that runs both, so this is the pair doing the same work on the same hardware.
| Mistral Small 4 (119B-2603) | Qwen3 30B-A3B Instruct 2507 | |
|---|---|---|
| Speed at 32k | 32 tok/s estimated | 25 tok/s measured |
| Pay-back on this machine | Pays back in 114 years | Pays back in 215 years |
| API cost per month | $2.71 | $1.57 |
Run Mistral Small 4 (119B-2603) on the Framework Desktop, 128GB · or Qwen3 30B-A3B Instruct 2507
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, 128GB. Mistral Small 4 (119B-2603) 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 | Mistral Small 4 (119B-2603) | Qwen3 30B-A3B Instruct 2507 |
|---|---|---|
| 50ka few chats a day | 1,140 years | 2,153 years |
| 200klight assistant use | 285 years | 538 years |
| 1Ma moderate coding-assistant day | 57 years | 108 years |
| 4Mheavy coding with an agent | 14 years | 27 years |
| 20Magents running most of the day | 2.8 years | 5.4 years |
Run Mistral Small 4 (119B-2603) at 20M tokens a day · or Qwen3 30B-A3B Instruct 2507
Machines that run one and not the other
Qwen3 30B-A3B Instruct 2507 needs 22 GB of memory at 32k of context and Mistral Small 4 (119B-2603) needs 75 GB. That puts Qwen3 30B-A3B Instruct 2507 on 16 of the 37 machines priced here that Mistral Small 4 (119B-2603) does not, starting at $1,269.
| Machine | Price | Memory | Speed on Qwen3 30B-A3B Instruct 2507 | Pay-back |
|---|---|---|---|---|
| Strix Halo Framework Desktop, 32GB | $1,269 | 32 GB | 27 tok/s estimated | Pays back in 78 years |
| AMD Radeon AI PRO R9700, 32GB | $1,299card only | 32 GB | 70 tok/s measured | Pays back in 79 years |
| Strix Halo Corsair AI Workstation 300, 64GB | $1,700 | 64 GB | 27 tok/s estimated | Pays back in 105 years |
| Strix Halo Framework Desktop, 64GB | $1,959 | 64 GB | 25 tok/s measured | Pays back in 122 years |
| NVIDIA GeForce RTX 5090, 32GB | $1,999card only | 32 GB | 104 tok/s measured | Pays back in 126 years |
| Strix Halo GMKtec EVO-X2, 64GB | $2,200 | 64 GB | 33 tok/s measured | Pays back in 134 years |
10 more, on the Qwen3 30B-A3B Instruct 2507 page.
Mistral Small 4 (119B-2603) is also head to head with GLM-4.5-Air above it on the leaderboard. One more model needs much the same memory: Ling 3.0 flash. Qwen3 30B-A3B Instruct 2507 is also head to head with Qwen3-Coder 30B-A3B, the current Qwen nearest it in size. One more model needs much the same memory: Nemotron 3.5 Bolt 74B-A9B.
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 Mistral Small 4 (119B-2603) · every machine that runs Qwen3 30B-A3B Instruct 2507 · every other match-up · both against the frontier · the quickest pay-back at each level of use