Apriel 1.6 15B Thinker — Hardware Requirements & GPU Compatibility
Vision
Apriel 1.6 15B Thinker is ServiceNow's 15-billion-parameter multimodal reasoning model, an update to Apriel-1.5-15B-Thinker that handles both text and images. It was post-trained with supervised fine-tuning and reinforcement learning, and the card reports a score of 57 on the Artificial Analysis index and a reduction of reasoning token usage of more than 30 percent compared with version 1.5. The card says it fits on a single GPU, and once quantized it is practical on a mainstream consumer card. The context window is 262,400 tokens. It is released under the MIT license, permitting unrestricted commercial and research use. Published in November 2025, it succeeds Apriel-1.5-15B-Thinker with a simplified chat template and four new special tokens to make tool calls and final responses easier to parse.
59.2K downloads
306 likesSpecifications
- Publisher
- ServiceNow-AI
- Parameters
- 14.9B
- Architecture
- LlavaForConditionalGeneration
- Context Length
- 262,400 tokens
- Vocabulary Size
- 131,072
- Release Date
- 2025-11-28
- License
- MIT
Get Started
HuggingFace
How Much VRAM Does Apriel 1.6 15B Thinker Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| Q2_K | 3.40 | 7.1 GB | 71.1 GB | 6.32 GB | 2-bit quantization with K-quant improvements |
| Q3_K_S | 3.50 | 7.3 GB | 71.3 GB | 6.50 GB | 3-bit small quantization |
| Q3_K_M | 3.90 | 8.1 GB | 72.0 GB | 7.25 GB | 3-bit medium quantization |
| Q4_0 | 4.00 | 8.2 GB | 72.2 GB | 7.43 GB | 4-bit legacy quantization |
| Q4_K_M | 4.80 | 9.7 GB | 73.7 GB | 8.92 GB | 4-bit medium quantization — most popular sweet spot |
| Q5_K_M | 5.70 | 11.4 GB | 75.4 GB | 10.59 GB | 5-bit medium quantization — good quality/size tradeoff |
| Q6_K | 6.60 | 13.1 GB | 77.0 GB | 12.26 GB | 6-bit quantization, very good quality |
| Q8_0 | 8.00 | 15.7 GB | 79.7 GB | 14.86 GB | 8-bit quantization, near-lossless |
Which GPUs Can Run Apriel 1.6 15B Thinker?
Q4_K_M · 9.7 GBApriel 1.6 15B Thinker (Q4_K_M) requires 9.7 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 13+ GB is recommended. Using the full 262K context window can add up to 64.0 GB, bringing total usage to 73.7 GB. 40 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti, NVIDIA GeForce RTX 3080 Ti.
Runs great
— Plenty of headroom
Decent
— Enough VRAM, may be tightWhich Devices Can Run Apriel 1.6 15B Thinker?
Q4_K_M · 9.7 GB49 devices with unified memory can run Apriel 1.6 15B Thinker, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, iPad Pro M5 13" (16 GB).
Runs great
— Plenty of headroomDecent
— Enough memory, may be tightWhere to Download Apriel 1.6 15B Thinker
Community quantizations of this model — GGUF for llama.cpp, Ollama, and LM Studio, plus AWQ/MLX variants where available.
Frequently Asked Questions
- How much VRAM does Apriel 1.6 15B Thinker need?
Apriel 1.6 15B Thinker requires 9.7 GB of VRAM at Q4_K_M, or 30.5 GB at BF16. Full 262K context adds up to 64.0 GB (73.7 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 14.9B × 4.8 bits ÷ 8 = 8.9 GB
KV Cache + Overhead ≈ 0.8 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 64.8 GB (at full 262K context)
VRAM usage by quantization
Q4_K_M9.7 GBQ4_K_M + full context73.7 GB- Can NVIDIA GeForce RTX 4090 run Apriel 1.6 15B Thinker?
Yes, at Q8_0 (15.7 GB) or lower. Higher quantizations like BF16 (30.5 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.
- What's the best quantization for Apriel 1.6 15B Thinker?
For Apriel 1.6 15B Thinker, Q4_K_M (9.7 GB) offers the best balance of quality and VRAM usage. Q4_K_L (9.9 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XXS at 4.9 GB.
VRAM requirement by quantization
IQ2_XXS4.9 GBIQ3_XS6.9 GBQ3_K_L8.4 GBQ4_K_M ★9.7 GBQ4_K_L9.9 GBBF1630.5 GB★ Recommended — best balance of quality and VRAM usage.
- Can I run Apriel 1.6 15B Thinker on a Mac?
Apriel 1.6 15B Thinker requires at least 4.9 GB at IQ2_XXS, which exceeds the unified memory of most consumer Macs. You would need a Mac Studio or Mac Pro with a high-memory configuration.
- Can I run Apriel 1.6 15B Thinker locally?
Yes — Apriel 1.6 15B Thinker can run locally on consumer hardware. At Q4_K_M quantization it needs 9.7 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is Apriel 1.6 15B Thinker?
At Q4_K_M, Apriel 1.6 15B Thinker can reach ~494 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~67 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B200 → 8000 ÷ 9.7 × 0.65 = ~535 tok/s
Estimated speed at Q4_K_M (9.7 GB)
~535 tok/s~67 tok/s~535 tok/s~494 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of Apriel 1.6 15B Thinker?
At Q4_K_M, the download is about 8.92 GB. The full-precision BF16 version is 29.73 GB. The smallest option (IQ2_XXS) is 4.09 GB.
- Which GPUs can run Apriel 1.6 15B Thinker?
40 consumer GPUs can run Apriel 1.6 15B Thinker at Q4_K_M (9.7 GB). Top options include AMD Radeon RX 6800, AMD Radeon RX 6800 XT, AMD Radeon RX 6900 XT, AMD Radeon RX 6700 XT. 26 GPUs have plenty of headroom for comfortable inference.
- Which devices can run Apriel 1.6 15B Thinker?
52 devices with unified memory can run Apriel 1.6 15B Thinker at Q4_K_M (9.7 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Apple iPhone 17 Pro, Asus ROG Flow Z13 (2025, Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.