Apriel-v1.5-15B-Thinker
Intelligence
28.3
#177/523
Coding
18.7
#222/429
Math
87.5
#42/265
Speed
104 tok/s
TTFT: 178.00s
Pricing
—
Apriel-v1.5-15B-Thinker is ServiceNow’s advanced AI model designed for complex problem-solving and data analysis. It processes at a speed of 104.114 tokens per second, making it suitable for professional users seeking efficient computational capabilities.
When to Use Apriel-v1.5-15B-Thinker
✓ Best For
- Data analysis and reporting.
- Mathematical problem solving.
- Automating coding tasks.
✗ Not Ideal For
- High-speed coding applications due to lower coding index.
- General conversational tasks where creativity is prioritized.
How Apriel-v1.5-15B-Thinker Compares
Intelligence Index · Higher is better
DeepSeekAlibabaServiceNowAmazon
Benchmark Profile
Coding Index
xAIInclusionAIServiceNowOpenAI
Math Index
GoogleDeepSeekServiceNowZ AI
Intelligence · Coding · Math
Intelligence
Coding
Math
All Benchmark Scores (13)
| Benchmark | Score | |
|---|---|---|
| Intelligence Index | 28.3 | |
| Coding Index | 18.7 | |
| Math Index | 87.5 | |
| MMLU-Pro | 773% | |
| GPQA | 713% | |
| LiveCodeBench | 728% | |
| HLE | 12% | |
| SciCode | 34.8% | |
| IFBench | 61.7% | |
| LCR | 20% | |
| TerminalBench Hard | 10.6% | |
| Tau2 | 68.4% | |
| AIME 2025 | 87.5% |
Data: Artificial Analysis · Updated: April 2, 2026
Frequently Asked Questions (15)
When was Apriel-v1.5-15B-Thinker released?
Apriel-v1.5-15B-Thinker was released on September 30, 2025.
Who created Apriel-v1.5-15B-Thinker?
Apriel-v1.5-15B-Thinker was created by ServiceNow.
How intelligent is Apriel-v1.5-15B-Thinker?
Apriel-v1.5-15B-Thinker scores 28 (estimated) on the Artificial Analysis Intelligence Index, placing it well above average among other open weight models of similar size (median: 15).
How fast is Apriel-v1.5-15B-Thinker?
Apriel-v1.5-15B-Thinker generates output at 139.4 tokens per second (based on the median across providers serving the model), which is above average compared to other open weight models of similar size (median: 98.0 t/s).
What is the latency of Apriel-v1.5-15B-Thinker?
Apriel-v1.5-15B-Thinker has a time to first token (TTFT) of 0.39s (based on the median across providers serving the model), which is very competitive compared to other open weight models of similar size (median: 1.77s).
How verbose is Apriel-v1.5-15B-Thinker?
When evaluated on the Intelligence Index, Apriel-v1.5-15B-Thinker generated 130M output tokens, which is at the higher end compared to other open weight models of similar size (median: 20M).
Is Apriel-v1.5-15B-Thinker a reasoning model?
Yes, Apriel-v1.5-15B-Thinker is a reasoning model. It uses extended thinking or chain-of-thought reasoning to work through complex problems before providing an answer.
What input modalities does Apriel-v1.5-15B-Thinker support?
Apriel-v1.5-15B-Thinker supports text and image input.
What output modalities does Apriel-v1.5-15B-Thinker support?
Apriel-v1.5-15B-Thinker supports text output.
Can Apriel-v1.5-15B-Thinker process images?
Yes, Apriel-v1.5-15B-Thinker supports image input and can analyze, describe, and answer questions about images.
Is Apriel-v1.5-15B-Thinker multimodal?
Yes, Apriel-v1.5-15B-Thinker is multimodal. It can process text and image input and generate text output.
What is the context window of Apriel-v1.5-15B-Thinker?
Apriel-v1.5-15B-Thinker has a context window of 130k tokens. This determines how much text and conversation history the model can process in a single request.
Is Apriel-v1.5-15B-Thinker open source?
Yes, Apriel-v1.5-15B-Thinker is open weights. The model weights are publicly available and can be downloaded for self-hosting.
How many parameters does Apriel-v1.5-15B-Thinker have?
Apriel-v1.5-15B-Thinker has 15 billion parameters.
What is the license for Apriel-v1.5-15B-Thinker?
Apriel-v1.5-15B-Thinker is released under the MIT license. This license allows commercial use.
