What hardware do you need to run Qwen3 8B?

Qwen3 8B at Q4_K_M is 5.0 GB of weights, with a context ceiling of 40k tokens. Quick and competent on short tasks. Reasoning mode helps a little and costs a lot of tokens. On the cheapest machine that runs it, a Mac mini M6, 16GB, it generates 12 tok/s at 32k of context, worked out from memory bandwidth.

How good is Qwen3 8B, really?

On the Artificial Analysis Intelligence Index v4.3 it scores 5 (reasoning mode; 6 without), which puts it in the Below every hosted tier band. Fine for simple, well-specified tasks. Noticeably less capable than anything the big labs sell today. See the whole table.

• Summarising — good
• Translation — usable
• Everyday coding — usable
• Reasoning & maths — don’t
• Agentic work — don’t

Head to head with its neighbours on the leaderboard: vs Ministral 3 14B above it, vs Gemma 3 27B it below. Also head to head: vs Qwen3.5 9B, the current Qwen nearest it in size. Two more models need much the same memory at 32k of context: vs Granite 4.2 8B, vs Gemma 3 12B it.

Is Qwen3 8B still worth running?

Not on the two figures this site can compare. The intelligence index puts seven of the current models here above Qwen3 8B, and not one of them asks more of a machine: at 32k of context it needs 9.9 GB and scores 5 (reasoning mode; 6 without), where the strongest of the seven, Qwen3.5 9B, needs 6.8 GB and scores 14 (reasoning mode; 13 without).

The cheapest machine still sold that holds Qwen3 8B at 32k is the Mac mini M6, 16GB at $899, which holds Qwen3.5 9B as well. This site quotes Qwen3.5 9B at 17 tok/s estimated there and Qwen3 8B at 12 tok/s estimated.

Qwen3.5 9B's score is Artificial Analysis's own estimate rather than a measured run of the full v4.3 suite.

What Qwen3 8B costs either way

Nobody rents Qwen3 8B by the token. The closest hosted match, Ministral 3 8B, costs $0.15 per million input tokens and $0.15 per million output (OpenRouter, cheapest active endpoint, checked 2026-10-29). Buying a machine only beats that if you use it hard enough, for long enough, that the hardware price divides down below the rental bill. What a million tokens costs each way puts the two prices side by side.

Machines that run Qwen3 8B

Every machine here holds it to 32k, the longest setting below this model's own 40k limit, so the choice between them is speed and price rather than how much you can keep in the window.

All 56 machines on this site run it at 32k, not just the eight in the table. One table sets every machine against the others: price, the memory its GPU can use, what it holds and how long it takes to pay for itself.

One machine per family, cheapest first. Speeds are measured where a public benchmark exists and estimated from memory bandwidth otherwise; the calculator says which for any configuration. The longest context is the longest setting the calculator offers that the machine still holds this model at, cache included, and no machine is shown taking it past its own 40k limit. Each one opens the calculator on that machine at that length.

For scale, the calculator starts from 80 tok/s for a hosted API and times a local machine against it. 1 of the 8 above reaches it, and the slowest is 12 tok/s.

The cheapest machine that runs it is sold with 16 GB. There is a page on what 16 GB runs, machine by machine: it names the 6 machines here sold at that size, and the 15 current models the roomiest of them holds at 32k.

It fits in less memory than that: the smallest size here that runs it at 32k is 12 GB, on the GeForce RTX 3060, 12GB at $329, card only, which hands a model 11 GB and is no longer sold.

What has been measured running Qwen3 8B

One of the speeds in the table above was measured, and there are more measurements than that table shows. Seven machines here have a published speed for Qwen3 8B. The table above has one of them; the other six are no longer sold. They are below, as published and again at 32k of context, which is what every other speed on this page is taken at.

A published figure is what its source reports, at the window the model was run at. The 32k figure is that measurement read at 32k of context: every token generated reads the weights and the whole key-value cache, so the longer the window the slower it runs. A discontinued machine's price is the one it launched at. Where two memory tiers of one machine carry the same figure, the speed is set by the bandwidth of the chip they share rather than by the amount of memory. Each published figure links the source that took it.

The specifics

Parameters

8.2B

Quantisation

Q4_K_M

Weights on disk

5.0 GB

KV cache

4.8 GB at 32k context. How weights and cache add up.

Maximum context

40k tokens (32k natively; 128k with YaRN, which llama.cpp can enable)

Licence

Apache 2.0

Sources

bartowski/Qwen_Qwen3-8B-GGUF, Qwen/Qwen3-8B