Qwen2.5 72B Instruct
Qwen · released Sep 16, 2024 · Qwen/Qwen2.5-72B-Instruct
- Type
- Open weightsCustom licence
- Params
- 72.7B
- Context
- 33K
about 25K words of context · download allowed, licence restricts use
Our take
Written Aug 3, 2026Qwen 2.5 is a 72.7-billion-parameter text-only instruction model released in September 2024. It carries a custom licence and offers broad Arena evaluation across six categories, though its graduate-level reasoning score sits near random-guess level.
Pick this for budget 72-billion-class inference, or for self-hosted deployment where custom licence terms are acceptable. Use it as an Arena-tested generalist with measured scores across coding, hard prompts, maths, instruction following and creative writing. Skip it if you need a permissive licence, if graduate-level reasoning is critical, or if you need guaranteed throughput — most offers lack speed data.
The case for it
- Broad Arena evaluation with six category Elos: coding 1355.5, hard prompts 1317.5, text overall 1302.9, maths 1296.4, instruction following 1292.3, and creative writing 1254.4.
- Strong instruction-following score of 84.8% on IFEval.
- Cheapest own-offer beats Novita by 5.3% on input price.
The case against it
- GPQA Diamond graduate-level reasoning score of 14.5% sits near random-guess level.
- Custom licence restricts commercial flexibility — not Apache or MIT, with restricted redistribution terms.
- Throughput unmeasured on three of five offers, and highly variable where known: 29 tokens per second versus 10 tokens per second.
How good is it?
IntelligencePuzzles, maths, exam questions
Arena Text (overall)125th of 143 · 1302.9
CodingWriting and fixing code on its own
Arena Coding117th of 143 · 1355.5
Arena Coding is the only board that has scored it for this.
AgenticPlanning, calling tools, staying on task
Nobody we watch has scored Qwen2.5 72B Instruct for this. We would take the rating from Arena Agent (IPS).
WritingWe do not rate this
Two boards come close and neither tests writing: Arena Creative Writing asks people which of two replies they prefer, and LiveBench Language tests whether a model understood a passage. So we show where Qwen2.5 72B Instruct placed and give it no mark out of five.
These tests check whether a model follows instructions — a precondition for all the work above, but not a measure of how well that work is done, which is why they get no rating.
Every published score for this model9 scoresEvery figure we hold, from 9 boards, with who ran it and a link to the source — including the boards no rating above is built on.
Can you run it yourself?
- Fits in memory
- weights load entirely on the card
- Spills to system RAM
- some weights offload; much slower
- Too large
- will not load even with offload
- est
- size is calculated; the verdict could change by 10%
GeForce RTX 4090 · 24 GB
Too large for this card. The weights do not fit even with part of them offloaded to system memory.
Apple M1 Pro (16-core GPU) · 32 GB
Too large for this card. The weights do not fit even with part of them offloaded to system memory.
Comfortable fit
Apple M2 Max (38-core GPU) · 96 GB
Room to spare. 22.9 GB spare means a 10% error in the size would not change the answer.
Memory use by level
Against a 24 GB card.
All 3 sizes here are calculated, not measured. We hold no measured file for this model, so each size comes from the parameter count and every verdict above inherits that uncertainty.
Check against your own machine → · All 71 devices, with every size →
Or rent it from someone else
Cheapest of 5 live listings. Picked at the widest standard context we hold, within one quantisation slice, so the numbers beside it are a price one host actually charges.
- per 1M tokens
- $0.36 in / $0.40 out
- Context served
- 33K
- Throughput
- Not measured
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| OpenRouter | $0.36 / $0.40 | 33K | not measured | Unknown | Unknown | Unknown |
| DeepInfrafp8 | $0.36 / $0.40 | 33K | not measured | Unknown | Unknown | Unknown |
| DeepInfrafp8 | $0.36 / $0.40 | 33K | 27 tok/s | No | No | Confirmed |
| Novita AI | $0.38 / $0.40 | 32K | not measured | Unknown | Unknown | Unknown |
| Novita AIbf16 | $0.38 / $0.40 | 32K | 10 tok/s | No | No | Confirmed |
DeepInfra
Novita AIAcross the 5 listings we hold: 2 say they do not train on prompts, 0 say they do and 3 do not say. 2 appear in the zero-retention registry we check; the rest are unknown to us rather than confirmed either way.
What each host's API supports
From the parameter list each endpoint publishes. Streaming is omitted: nothing we hold reports it, for any model.
- ✓
- Supported
- ✗
- Not supported
- Not published
- host gave no parameter list
| Provider | Tool calling | JSON output | Strict schema |
|---|---|---|---|
| OpenRouter | ✓ | ✓ | ✓ |
| DeepInfrafp8 | |||
| DeepInfrafp8 | ✓ | ✓ | ✓ |
| Novita AI | |||
| Novita AIbf16 | ✓ | ✓ | ✗ |
Tool calling: 3 of 5 listings say yes, 2 publish no parameter list. JSON output: 3 of 5 listings say yes, 2 publish no parameter list. Strict schema: 2 of 5 listings say yes, 1 says no, 2 publish no parameter list.
Models people weigh against Qwen2.5 72B Instruct
When we formed this view
Dates behind this page
Prices last checked 5d ago
What we do not know about this model yet
- We hold no measured file for it, so all 3 sizes on this page are calculated from the parameter count.
- 2 of 5 listings publish no parameter list, so what their API accepts is unknown to us.
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
- 3 of 5 listings do not say whether they train on prompts.
- We hold no cached-input rate for any of its listings.
Licence and identifiers
What the licence allowsCustom licence, what it allows commercially, and the identifiers you need to pull this model — its Hugging Face repo, our slug and a machine-readable card.
Licence
Custom licence
This model ships custom license terms that don't map to a known template. We haven't parsed them, so commercial use, redistribution and derivatives are unverified — review the original terms before shipping.
Identifiers
- Hugging Face
- Qwen/Qwen2.5-72B-Instruct
- Architecture
- Dense
- Modality record
- text->text
- Catalogue slug
- qwen-qwen2-5-72b-instruct