Leaderboard / Edge0
Edge0 35B-A3B (preview)
Edge0 · Edge0 · released 2026-09-08 · data checked 2026-11-01
Edge0 35B-A3B (preview) is a 34.7B-parameter Mixture-of-Experts model (3B active per token) from Edge0, released 2026-09-08 under the Apache-2.0 license. It scores 62.8 on the Small LM Index, #15 of 34 medium models. It needs about 22.4 GB of memory at 4-bit with an 8K-token context.
On-device build of Qwen3.6 35B-A3B. Scores were measured at int4; Edge0 claims ~3 GB peak memory on a Mac mini.
Benchmarks
Each bar shows this model's score. The tick marks the best medium model on that benchmark. Sources: 4 reported by Edge0. How we label sources.
Not reported: estimated
What the index model expects from this model's other scores. Not a measurement.
Specifications
- Lab
- Edge0 (country not published · Other)
- Based on
- Qwen3.6 35B-A3B (fine-tune or distillation)
- Architecture
- Mixture-of-Experts
- Total parameters
- 34.7B
- Active parameters
- 3B
- Size class
- Medium
- Context window
- 262K tokens
- Input modalities
- text, image
- Reasoning
- Hybrid (thinking on/off)
- License
- Apache-2.0
- Openness
- Open weights
- Memory to run
- ~22.4 GB at 4-bit, 8K context (more sizes)
- Hugging Face
- 4.2K downloads (30 days) · 1.1K likes
Compare with
- Edge0 35B-A3B (preview) vs Qwen3.6 35B-A3B (35B, index 72.2)
- Edge0 35B-A3B (preview) vs Qwen3.5 35B-A3B (35B, index 69.9)
- Edge0 35B-A3B (preview) vs MiniCPM-V 4.7 35B-A3B (35.2B, index 54.5)
- Edge0 35B-A3B (preview) vs Seed-OSS 36B Instruct (36.2B, index 60.3)
Sources
Something wrong here? Report it.
Run it locally
No GGUF build from a well-known publisher yet.
vllm serve Edge0/Edge0-35B-A3B-previewMemory needed: weights, context cache and runtime buffers.
| Quantization | 4K context | 32K context | 256K context |
|---|---|---|---|
| 16-bit F16 | 71 GB | 71 GB | 76 GB |
| 8-bit Q8_0 | 38 GB | 39 GB | 43 GB |
| 4-bit Q4_K_M | 22 GB | 23 GB | 28 GB |