Spark-X2.6 4B
Apache 2.0SparkLLM · 4.1B · Dense
On-device agent LLM with 1M context and a configurable thinking budget Check if your GPU or Mac can run Spark-X2.6 4B locally — 2.3 GB min, 3.8 GB recommended.
2026-101024K context
| Quant | Bits | VRAM | Quality | Status |
|---|---|---|---|---|
| Q2_K | 2 | 1.8 GB | low | — |
| Q3_K_M | 3 | 2.3 GB | moderate | — |
| Q4_K_M | 4 | 2.6 GB | good | — |
| Q5_K_M | 5 | 3.1 GB | good | — |
| Q6_K | 6 | 3.7 GB | excellent | — |
| Q8_0 | 8 | 4.7 GB | excellent | — |
| F16 | 16 | 8.9 GB | lossless | — |
Can I run Spark-X2.6 4B locally?
- Can I run Spark-X2.6 4B locally?
- Spark-X2.6 4B needs about 2.3 GB of memory at a minimum and 3.8 GB recommended. Open this page to grade it against your GPU or Mac, then run it with runai, Ollama or LM Studio.
- How much VRAM does Spark-X2.6 4B need?
- At Q4_K_M, Spark-X2.6 4B uses about 2.6 GB of VRAM. Higher quants need more memory; lower quants fit tighter cards with a quality tradeoff.