S

Apertus 8B Instruct

Sep 2025
SApertus 8B Instruct
Compare in
on
—per task?
$0.11 / $0.28per 1M in / out?
148 tok/sfirst token in 2.9 s?
66Kcontext

Ranking?

Strong at

C
Tool calling#221 of 242Reasoning#275 of 334Science#307 of 312Knowledge#305 of 318Instructions#235 of 245

Alternatives?

ModelScoreCostPer taskWhat changes
CheaperLing 3.0 Flash VLthinkingOpen×5.1-40%$0.06 / $0.18Memory (4-bit): 4 GB → 75 GBBetterGPT-6 Lunalow×4.5+33%$0.005Function calling only with none · Chat Completions+3IOpenK2 Horizon 7Bthinking×4.3+248%$0.30 / $1.20Memory (4-bit): 4 GB → 4.5 GBCheaperSolar Mini 4×5-42%$0.37CheaperLing 3.0 Flash FinthinkingOpen×4.7-40%$0.06 / $0.18Memory (4-bit): 4 GB → 76 GB

Run it yourself

estimate
Fits on✓ 8 GB (phone, laptop) · compressed✓ 16 GB · 8-bit✓ 24 GB GPU · full quality✓ 48 GB GPU · full quality✓ 80 GB GPU · full quality
LicenceApache-2.0free for commercial use

Benchmarks

all categories
Chat & Knowledge
Humanity's Last Exam5#277
AA Omniscience Index-77.2#308
CritPt (physics)0#180
AA Intelligence Index 4.3.2≈4.8#327
AA Omniscience · Accuracy9.7#282
AA Omniscience · Hallucination Rate96.1#314
GPQA Diamond25.6#310
Coding
Terminal-Bench Hard (version unstated)0#213
Agents & Tool Use
τ²-bench · AA11.4#224
Writing & Language
IFBench (instructions)22.4#238

FAQ

How much does Apertus 8B Instruct cost?›

$0.11 per 1M input tokens and $0.28 per 1M output tokens (cheapest host: Featherless).

Is Apertus 8B Instruct free to use?›

There is no free API; the cheapest paid price is $0.11 per 1M input tokens. The weights are open, so you can run it yourself.

How good is Apertus 8B Instruct, and what is it best at?›

#275 of 334 in Chat & Knowledge, #210 of 239 in Coding, #221 of 242 in Agents & Tool Use, #235 of 245 in Writing & Language.

Is Apertus 8B Instruct better than Ministral 3 3B 2512?›

On AA Intelligence Index 4.3.2, Apertus 8B Instruct scores 4.8 vs 4.8 — about the same.

How can I use Apertus 8B Instruct?›

Through the API of Featherless.

What is the context window of Apertus 8B Instruct?›

66K tokens.

When was Apertus 8B Instruct released? Is there a newer version?›

Released 2025-09-02. It is the current version in its line.

Is Apertus 8B Instruct open source?›

Yes, the weights are open (Apache-2.0 licence).

Can I run Apertus 8B Instruct locally?›

The weights are open and can be self-hosted; hardware needs are not confirmed yet.

Can Apertus 8B Instruct read images?›

No, it accepts text only.

What is Apertus 8B Instruct?›

Apertus 8B Instruct is an open-weights language model from Swiss AI Initiative, released Sep 2025.

What are the best alternatives to Apertus 8B Instruct?›

Ling 3.0 Flash VL (cheaper), GPT-6 Luna (better), K2 Horizon 7B (open weights).