Apertus 70B Instruct API Pricing
Swiss AI Initiative · released 2025-09-02
Pricing
0.082¢ / 1K
0.292¢ / 1K
0.135¢ / 1K
What it actually costs
| Workload | Tokens | Cost |
|---|---|---|
| Short chat reply | 500 in / 300 out | 0.129¢ |
| RAG query | 4,000 in / 500 out | 0.474¢ |
| Long document summary | 50,000 in / 1,000 out | $0.0439 |
| Code review of a 2k-line file | 30,000 in / 2,000 out | $0.0304 |
| 1,000 chat replies/day for 30 days | 30,000 × (500 in / 300 out) | $38.58 |
Cost = (input tokens ÷ 1,000,000 × $0.82) + (output tokens ÷ 1,000,000 × $2.92).
Benchmarks
#501 of 533
Similar intelligence, lower price
| Model | Intelligence | Blended $/1M |
|---|---|---|
| Llama 3.1 Instruct 8B | 6.9 | $0.028 |
| Sarvam 30B (high) | 6.6 | $0.047 |
| Nova Micro | 5.9 | $0.061 |
| Granite 4.1 8B | 6.6 | $0.063 |
| NVIDIA Nemotron Nano 9B V2 (Reasoning) | 7.4 | $0.07 |
Leaderboard neighbours
| Model | Intelligence | Blended $/1M |
|---|---|---|
| LFM2.5-1.2B-Thinking | 5.2 | — |
| Olmo 3 7B Instruct | 5.2 | $0.125 |
| Qwen3 1.7B (Reasoning) | 5.2 | — |
| Apertus 70B Instruct | 5.1 | $1.345 |
| DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning) | 5.1 | — |
| Gemma 3 270M | 5.1 | — |
| Granite 4.0 Micro | 5.1 | — |
More from Swiss AI Initiative
| Model | Intelligence | Blended $/1M |
|---|---|---|
| Apertus 8B Instruct | 4.8 | $0.125 |
Count tokens & estimate cost for Apertus 70B Instruct See Apertus 70B Instruct on the leaderboard
Source: Artificial Analysis · last synced 2026-09-22
FAQ
Who develops Apertus 70B Instruct?
Apertus 70B Instruct is developed by Swiss AI Initiative, released 2025-09-02.
How much does Apertus 70B Instruct cost per million tokens?
Apertus 70B Instruct costs $0.82 per 1M input tokens and $2.92 per 1M output tokens ($1.345 blended), per Artificial Analysis's official pricing data.
How does Apertus 70B Instruct rank on benchmarks?
Apertus 70B Instruct scores 5.1 on the Artificial Analysis Intelligence Index, ranking #501 of 533 models we track.
Is there a cheaper model with similar intelligence to Apertus 70B Instruct?
Yes — Llama 3.1 Instruct 8B scores similar intelligence (6.9) at $0.028/1M blended, versus $1.345/1M for Apertus 70B Instruct.
Where does this data come from?
Pricing and benchmark data for Apertus 70B Instruct comes from Artificial Analysis (artificialanalysis.ai), synced daily. Last synced 2026-09-22.