GLM-4.5-Air vs Devstral 2 123B
GLM-4.5-Air scores higher on the intelligence index, 11 against 9. Both take the same machine to start: the cheapest here that runs either is the Framework Desktop, 128GB, at $3,449. On it, GLM-4.5-Air is about 4.5× quicker: 10 tok/s against 2.2, both estimated from memory bandwidth. At 500k tokens a day the Framework Desktop, 128GB pays for itself in 59 years running Devstral 2 123B, against 139 years running GLM-4.5-Air.
| GLM-4.5-Air | Devstral 2 123B | |
|---|---|---|
| Intelligence index | 11 | 9 |
| Class | Below every hosted tier | Below every hosted tier |
| Weights | 73 GB | 75 GB |
| Needs at 32k | 79 GB | 87 GB |
| Quantisation | Q4_K_M | Q4_K_M |
| Parameters | 110.5B (12B active) | 125B |
| Max context | 128k | 256k |
| API price per 1M | $0.13 in / $0.85 out | $0.4 in / $2 out |
| Licence | MIT | other (see repo) |
| Machines here that run it | 12 of 37 | 12 of 37 |
| Cheapest machine that runs it | Strix Halo Framework Desktop, 128GB $3,449 | Strix Halo Framework Desktop, 128GB $3,449 |
Run GLM-4.5-Air on the Framework Desktop, 128GB · or Devstral 2 123B
Jump to: Side by side on the Framework Desktop, 128GB · 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, 128GB
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, 128GB, at $3,449.
| GLM-4.5-Air | Devstral 2 123B | |
|---|---|---|
| Speed at 32k | 10 tok/s estimated | 2.2 tok/s estimated |
| Pay-back on this machine | Pays back in 139 years | Pays back in 59 years |
| API cost per month | $2.66 | $7.61 |
Run GLM-4.5-Air on the Framework Desktop, 128GB · or Devstral 2 123B
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. Devstral 2 123B pays for it sooner at every level up to 4M tokens a day. Above that GLM-4.5-Air does.
| A day's use | GLM-4.5-Air | Devstral 2 123B |
|---|---|---|
| 50ka few chats a day | 1,392 years | 585 years |
| 200klight assistant use | 348 years | 146 years |
| 1Ma moderate coding-assistant day | 70 years | 29 years |
| 4Mheavy coding with an agent | 17 years | 9.6 yearsits ceiling |
| 20Magents running most of the day | 5.0 yearsits ceiling | 9.6 yearsits ceiling |
The Framework Desktop, 128GB generates at most 13.8M tokens a day on GLM-4.5-Air, so that column's figure at 20M tokens a day is for the most it can do, not for the whole of what was asked. It generates at most 3.06M tokens a day on Devstral 2 123B, so that column's figures from 4M tokens a day up are for the most it can do, not for the whole of what was asked.
Run GLM-4.5-Air at 20M tokens a day · or Devstral 2 123B
Memory is not what separates them
GLM-4.5-Air needs 79 GB of memory at 32k of context and Devstral 2 123B needs 87 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.
GLM-4.5-Air is also head to head with Granite 4.2 8B above it on the leaderboard and Mistral Small 4 (119B-2603) below it. One more model needs much the same memory: Qwen3.5 122B-A10B. Devstral 2 123B is also head to head with Gemma 4 E4B above it on the leaderboard and DeepSeek-R1-Distill-Qwen-32B below 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 GLM-4.5-Air · every machine that runs Devstral 2 123B · every other match-up · both against the frontier · the quickest pay-back at each level of use