Granite 4.2 30B vs GLM-4.7-Flash
Granite 4.2 30B and GLM-4.7-Flash both score 15 on the intelligence index. The cheapest machine here that runs Granite 4.2 30B is the Radeon AI PRO R9700, 32GB, at $1,299. GLM-4.7-Flash runs on the Framework Desktop, 32GB at $1,269, $30 less. The Radeon AI PRO R9700, 32GB is priced as the card alone, without the PC around it. On the Radeon AI PRO R9700, 32GB, the cheapest machine here that runs both, GLM-4.7-Flash is about 4.5× quicker: 94 tok/s against 21, both estimated from memory bandwidth. At 500k tokens a day the Radeon AI PRO R9700, 32GB pays for itself in 98 years running GLM-4.7-Flash, against 178 years running Granite 4.2 30B.
| Granite 4.2 30B | GLM-4.7-Flash | |
|---|---|---|
| Intelligence index | 15 | 15 |
| Class | Haiku-class | Haiku-class |
| Weights | 18 GB | 18 GB |
| Needs at 32k | 26 GB | 20 GB |
| Quantisation | Q4_K_M | Q4_K_M |
| Parameters | 29.3B | 31.2B (3B active) |
| Max context | 128k | 198k |
| API price per 1M | $0.0605 in / $0.4 outpriced as GLM-4.7-Flash | $0.0605 in / $0.4 out |
| Licence | Apache 2.0 | MIT |
| Machines here that run it | 27 of 37 | 30 of 37 |
| Cheapest machine that runs it | AMD Radeon AI PRO R9700, 32GB $1,299card only | Strix Halo Framework Desktop, 32GB $1,269 |
Run Granite 4.2 30B on the Radeon AI PRO R9700, 32GB · or GLM-4.7-Flash on the Framework Desktop, 32GB
Jump to: Side by side on the Radeon AI PRO R9700, 32GB · 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 Radeon AI PRO R9700, 32GB
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 AMD Radeon AI PRO R9700, 32GB is the cheapest machine here that runs both, so this is the pair doing the same work on the same hardware.
| Granite 4.2 30B | GLM-4.7-Flash | |
|---|---|---|
| Speed at 32k | 21 tok/s estimated | 94 tok/s estimated |
| Pay-back on this machine | Pays back in 178 years | Pays back in 98 years |
| API cost per month | $1.24priced as GLM-4.7-Flash | $1.24 |
Run Granite 4.2 30B on the Radeon AI PRO R9700, 32GB · or GLM-4.7-Flash
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 Radeon AI PRO R9700, 32GB. GLM-4.7-Flash 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 | Granite 4.2 30B | GLM-4.7-Flash |
|---|---|---|
| 50ka few chats a day | 1,784 years | 984 years |
| 200klight assistant use | 446 years | 246 years |
| 1Ma moderate coding-assistant day | 89 years | 49 years |
| 4Mheavy coding with an agent | 22 years | 12 years |
| 20Magents running most of the day | 4.5 years | 2.5 years |
Run Granite 4.2 30B at 20M tokens a day · or GLM-4.7-Flash
Machines that run one and not the other
GLM-4.7-Flash needs 20 GB of memory at 32k of context and Granite 4.2 30B needs 26 GB. That puts GLM-4.7-Flash on 3 of the 37 machines priced here that Granite 4.2 30B does not, starting at $1,269.
| Machine | Price | Memory | Speed on GLM-4.7-Flash | Pay-back |
|---|---|---|---|---|
| Strix Halo Framework Desktop, 32GB | $1,269 | 32 GB | 40 tok/s estimated | Pays back in 97 years |
| Mac mini M6, 32GB | $1,299 | 32 GB | 14 tok/s estimated | Pays back in 104 years |
| MacBook Pro M5 (14-inch), 32GB | $2,399 | 32 GB | 13 tok/s estimated | Pays back in 196 years |
Granite 4.2 30B is also head to head with Gemma 4 31B it above it on the leaderboard. One more model needs much the same memory: Nemotron 3.5 Lightning 30B-A3B. GLM-4.7-Flash is also head to head with Gemma 4 12B below it on the leaderboard. One more model needs much the same memory: Devstral Small 2 24B.
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 Granite 4.2 30B · every machine that runs GLM-4.7-Flash · every other match-up · both against the frontier · the quickest pay-back at each level of use