Qwen2.5 7B Instruct
Qwen · released Sep 16, 2024 · Qwen/Qwen2.5-7B-Instruct
- Type
- Open weightsApache License 2.0
- Params
- 7.6B
- Context
- 33K
about 25K words of context
Our take
Written Sep 2, 2026Qwen 2.5 is a compact downloadable text model with a permissive Apache licence and a 32,768-token request limit. It scores well on instruction-following tests for its size and is among the cheapest hosted options we track, though its science reasoning and broad knowledge are limited.
Pick this for tight-budget text tasks where following instructions matters more than deep reasoning or encyclopaedic knowledge. Use it for self-hosting with commercial freedom, or for high-volume hosted inference where cost is the main constraint. Skip it if you need graduate-level science accuracy, broad domain mastery, or guaranteed fast throughput — speed varies more than twofold between providers.
The case for it
- Apache 2.0 licence allows commercial use, fine-tuning and redistribution without restriction.
- IFEval instruction-following accuracy of 72.8% is strong for a model of this scale.
- Hosted inference is very cheap: the cheapest tracked offer costs several times less than the most expensive one.
The case against it
- GPQA Diamond science reasoning at 6.5% correct is near the bottom of measured models.
- MMLU-Pro broad knowledge at 37.8% correct shows limited depth across domains.
- Throughput varies sharply by provider, with more than a twofold spread between endpoints.
How good is it?
EverydayGeneral questions and everyday reasoning
Not yet scored on Arena Text (overall). It is on GPQA Diamond, in 11th of 16 with 29.1.
CodingWriting and fixing code on its own
Not yet scored on Arena Coding.
AgenticPlanning, calling tools, staying on task
Not yet scored on Arena Agent.
WritingDrafting and rewriting prose
Not yet scored on Arena Creative Writing.
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.
Every published score for this model3 scoresEvery figure we hold, from 3 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?
Comfortable fit
GeForce RTX 4090 · 24 GB
Room to spare. 16.6 GB spare means a 10% error in the size would not change the answer.
GeForce RTX 5090 · 32 GB
Room to spare. 24.6 GB spare means a 10% error in the size would not change the answer.
Apple M1 (8-core GPU) · 16 GB
Room to spare. 5.8 GB spare means a 10% error in the size would not change the answer.
Memory use by level
Against a 24 GB card.
What is quantisation? →This model on every device we track71 devicesThe Q4 build most people download, on each device: what the weights come to, how much context the memory leaves, and whether it runs. Smallest device that runs it first.
Check against your own machine → · Where to rent it hosted →
Or rent it from someone else
Prices checked between 2 hours and 21 days ago — each listing carries its own date.
Phala, through OpenRouter
Cheapest of the 2 listings we can compare like for like — at 33K of context, out of 3 in the table below. One cheaper row there is outside that comparison: a different context length.
- per 1M tokens
- $0.10 in / $0.20 out
- Context served
- 33K
- Throughput
- ~28 tok/s
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| Novita AIDirect | $0.070 / $0.070checked 21 days ago | 32K | not measured | Unknown | Unknown | Unknown |
| OpenRouterOpenRouter's own listing | $0.10 / $0.20checked 2 hours ago | 33K | not measured | Unknown | Unknown | Unknown |
| PhalaThrough OpenRouter | $0.10 / $0.20checked 2 hours ago | 33K29K max reply | 28 tok/s | No | No | Confirmed |
Novita AI
OpenRouterAcross the 3 listings we hold: 1 says it does not train on prompts, 0 say they do and 2 do not say. 1 appears in the zero-retention registry we check; the rest are unknown to us.
What each host's API supports
From the parameter list each endpoint publishes. Streaming is omitted: nothing we hold reports it, for any model.
| Provider | Tool calling | JSON output | Strict schema |
|---|---|---|---|
| Novita AIDirect | |||
| OpenRouterOpenRouter's own listing | ✓ | ✓ | ✓ |
| PhalaThrough OpenRouter | ✓ | ✓ | ✓ |
Tool calling: 2 of 3 listings say yes, 1 publishes no parameter list. JSON output: 2 of 3 listings say yes, 1 publishes no parameter list. Strict schema: 2 of 3 listings say yes, 1 publishes no parameter list.
Models people weigh against Qwen2.5 7B Instruct
When we formed this view
Recent changes
What moved
input +150% ($0.040 → $0.100 per 1M tokens), output +100% ($0.10 → $0.20 per 1M tokens)What moved
first indexed by our pipelineEach date is the day we first saw the change, or the day the maker announced it.
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.
- 1 of 3 listings publishes no parameter list, so what its API accepts is unknown to us.
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
- 2 of 3 listings do not say whether they train on prompts.
- We hold no cached-input rate for any of its listings.
- We hold no batch or off-peak rate for any of its listings.
- We hold a decode speed for it, but no prompt-processing (prefill) figure, so how long the input side of a job takes is unknown to us.
Licence and identifiers
What the licence allowsApache License 2.0, 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
Apache License 2.0
Fully permissive: commercial use, redistribution, and derivatives allowed. Requires attribution and a copy of the license. Includes an express patent grant.
Identifiers
- Hugging Face
- Qwen/Qwen2.5-7B-Instruct
- Architecture
- Dense
- Takes in, gives back
- Text in, text out
- Catalogue slug
- qwen-qwen2-5-7b-instruct