Apriel-v1.6-15B-Thinker
Intelligence
27.6
#181/521
Coding
22
#196/427
Math
88
#38/265
Speed
139 tok/s
TTFT: 244.00s
Pricing
—
Apriel-v1.6-15B-Thinker is ServiceNow’s advanced AI model designed for various analytical tasks. It processes at a speed of 139.066 tokens per second and is priced at $/M input, $/M output, targeting professional users.
When to Use Apriel-v1.6-15B-Thinker
✓ Best For
- Data analysis and reporting.
- Mathematical problem solving.
- Coding assistance and debugging.
✗ Not Ideal For
- Tasks requiring high-level creative writing.
- Applications needing real-time responses.
How Apriel-v1.6-15B-Thinker Compares
Intelligence Index · Higher is better
MistralDeepSeekServiceNowAlibabaOpenAI
Benchmark Profile
Coding Index
AlibabaxAIServiceNowDeepSeekOpenAI
Math Index
AlibabaAnthropicServiceNowPrime IntellectGoogle
Intelligence · Coding · Math
Intelligence
Coding
Math
All Benchmark Scores (13)
| Benchmark | Score | |
|---|---|---|
| Intelligence Index | 27.6 | |
| Coding Index | 22 | |
| Math Index | 88 | |
| MMLU-Pro | 79% | |
| GPQA | 733% | |
| LiveCodeBench | 807% | |
| HLE | 98% | |
| SciCode | 37.3% | |
| IFBench | 69.1% | |
| LCR | 50.3% | |
| TerminalBench Hard | 14.4% | |
| Tau2 | 69.3% | |
| AIME 2025 | 88% |
Data: Artificial Analysis · Updated: March 27, 2026
Frequently Asked Questions (15)
When was Apriel-v1.6-15B-Thinker released?
Apriel-v1.6-15B-Thinker was released on November 25, 2025.
Who created Apriel-v1.6-15B-Thinker?
Apriel-v1.6-15B-Thinker was created by ServiceNow.
How intelligent is Apriel-v1.6-15B-Thinker?
Apriel-v1.6-15B-Thinker scores 28 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-v1.6-15B-Thinker?
Apriel-v1.6-15B-Thinker generates output at 128.2 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: 97.3 t/s).
What is the latency of Apriel-v1.6-15B-Thinker?
Apriel-v1.6-15B-Thinker has a time to first token (TTFT) of 0.44s (based on the median across providers serving the model), which is very competitive compared to other open weight models of similar size (median: 1.97s).
How verbose is Apriel-v1.6-15B-Thinker?
When evaluated on the Intelligence Index, Apriel-v1.6-15B-Thinker generated 75M output tokens, which is at the higher end compared to other open weight models of similar size (median: 19M).
Is Apriel-v1.6-15B-Thinker a reasoning model?
Yes, Apriel-v1.6-15B-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-v1.6-15B-Thinker support?
Apriel-v1.6-15B-Thinker supports text and image input.
What output modalities does Apriel-v1.6-15B-Thinker support?
Apriel-v1.6-15B-Thinker supports text output.
Can Apriel-v1.6-15B-Thinker process images?
Yes, Apriel-v1.6-15B-Thinker supports image input and can analyze, describe, and answer questions about images.
Is Apriel-v1.6-15B-Thinker multimodal?
Yes, Apriel-v1.6-15B-Thinker is multimodal. It can process text and image input and generate text output.
What is the context window of Apriel-v1.6-15B-Thinker?
Apriel-v1.6-15B-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-v1.6-15B-Thinker open source?
Yes, Apriel-v1.6-15B-Thinker is open weights. The model weights are publicly available and can be downloaded for self-hosting.
How many parameters does Apriel-v1.6-15B-Thinker have?
Apriel-v1.6-15B-Thinker has 15 billion parameters.
What is the license for Apriel-v1.6-15B-Thinker?
Apriel-v1.6-15B-Thinker is released under the MIT license. This license allows commercial use.
