Sunk Cost sunkcost.ai Data checked 2026-09-03

GLM-4.7-Flash vs Qwen3 30B-A3B Instruct 2507

GLM-4.7-Flash scores higher on the intelligence index, 15 against 10. Both take the same machine to start: the cheapest here that runs either is the Framework Desktop, 32GB, at $1,269. On it, GLM-4.7-Flash is about 1.5× quicker: 40 tok/s against 27, both estimated from memory bandwidth. At 500k tokens a day the Framework Desktop, 32GB pays for itself in 96 years running GLM-4.7-Flash, against 164 years running Qwen3 30B-A3B Instruct 2507.

GLM-4.7-FlashQwen3 30B-A3B Instruct 2507
Intelligence index1510
ClassHaiku-classBelow every hosted tier
Weights18 GB19 GB
Needs at 32k20 GB22 GB
QuantisationQ4_K_MQ4_K_M
Parameters31.2B (3B active)30.5B (3.3B active)
Max context198k256k
API price per 1M$0.061 in / $0.4 out$0.048 in / $0.193 out
LicenceMITApache 2.0
Machines here that run it30 of 3728 of 37
Cheapest machine that runs itStrix Halo Framework Desktop, 32GB $1,269Strix Halo Framework Desktop, 32GB $1,269
Summarising not rated good
Translation not rated good
Everyday coding not rated good
Reasoning & maths not rated usable
Agentic work not rated usable

Run GLM-4.7-Flash on the Framework Desktop, 32GB · or Qwen3 30B-A3B Instruct 2507

Ratings are coarse on purpose: they say what a model is usable for, not where it places to the decimal.

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.

GLM-4.7-FlashQwen3 30B-A3B Instruct 2507
Speed at 32k40 tok/s estimated27 tok/s estimated
Pay-back on this machinePays back in 96 yearsPays back in 164 years
API cost per month$1.25$0.87

Run GLM-4.7-Flash on the Framework Desktop, 32GB · 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, 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 useGLM-4.7-FlashQwen3 30B-A3B Instruct 2507
50ka few chats a day961 years1,635 years
200klight assistant use240 years409 years
1Ma moderate coding-assistant day48 years82 years
4Mheavy coding with an agent12 years20 years
20Magents running most of the day2.4 years4.1 years

Run GLM-4.7-Flash at 20M tokens a day · or Qwen3 30B-A3B Instruct 2507

Machines that run one and not the other

GLM-4.7-Flash needs 20 GB of memory at 32k of context and Qwen3 30B-A3B Instruct 2507 needs 22 GB. That puts GLM-4.7-Flash on 2 of the 37 machines priced here that Qwen3 30B-A3B Instruct 2507 does not, starting at $1,299.

MachinePriceMemorySpeed on GLM-4.7-FlashPay-back
Mac mini M6, 32GB$1,29932 GB14 tok/s estimatedPays back in 103 years
MacBook Pro M5 (14-inch), 32GB$2,39932 GB13 tok/s estimatedPays back in 195 years

GLM-4.7-Flash is also head to head with Granite 4.2 30B above it on the leaderboard and Gemma 4 12B below it. Two more models need much the same memory: Devstral Small 2 24B and Qwen3-Coder 30B-A3B. Qwen3 30B-A3B Instruct 2507 is also head to head with Mistral Small 4 (119B-2603) above it on the leaderboard and Qwen3-Coder 30B-A3B, the current Qwen nearest it in size.

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-4.7-Flash · 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