Cola2
Cola2 is an open-weights language model from ByteDance, released November 6, 2026. It has 8.8B parameters (dense). The context window is 32K tokens.
Check it against your own PC →
Can you run Cola2 locally?
The smallest common GPU it runs on is 6 GB of VRAM (with 32 GB of RAM): Q6_K, split GPU / CPU, roughly 5 tokens per second. Estimates assume 32 GB of DDR5 RAM, a 16K context and llama.cpp-style GGUF quantization. Enter your exact hardware in the live app.
| GPU memory | Verdict | Best quant | VRAM | RAM | tok/s |
|---|---|---|---|---|---|
| 6 GB VRAM | Runs, slowly | Q6_K (≈ lossless) | 5.2 GB | 4.8 GB | ~5 |
| 8 GB VRAM | Runs on GPU | Q3_K_M (okay) | 7.1 GB | – | ~37 |
| 12 GB VRAM | Runs on GPU | Q6_K (≈ lossless) | 10.0 GB | – | ~29 |
| 16 GB VRAM | Runs on GPU | Q6_K (≈ lossless) | 10.0 GB | – | ~40 |
| 24 GB VRAM | Runs on GPU | Q6_K (≈ lossless) | 10.0 GB | – | ~72 |
| 32 GB VRAM | Runs on GPU | Q6_K (≈ lossless) | 10.0 GB | – | ~130 |
Benchmark results
Epoch AI has not published results for Cola2 yet. New models are usually evaluated within days to weeks.
Links
Hugging Face: ByteDance-Seed/Cola2
More from ByteDance
Doubao Seed 2.1 Turbo · Doubao Seed 2.1 Pro · Doubao Seed Evolving (Nov 2026) · Doubao Seed 2.0 Lite (Apr 2026) · Doubao-Seed-2.0 Pro · Doubao Seed 2.0 Code · Doubao Seed 2.0 Mini · Doubao Seed 1.8