Can T4 16GB run Solar Open 2?

NVIDIA · 16 GB GDDR6 · 320 GB/s bandwidth · Upstage · 250.0B (active 15.0B) · MoE · model ctx up to 1024K

Fit & speed by quant and context

Fits = weights + KV(ctx) + 1 GB overhead ≤ 95% of VRAM. Measured anchors are context-agnostic; the fit verdict is recomputed per context. A missing tok/s means the model is far beyond this card (offload-only territory).

VRAM breakdown at 8K context

Other GPUs that run Solar Open 2

• B200 192GB SXM — ≈510.3 tok/s Q4 est.
• MI300X 192GB — ≈333.6 tok/s Q4 est.
• H200 141GB SXM5 — ≈306.2 tok/s Q4 est.
• DGX Spark (GH200 480GB) — ≈30.9 tok/s Q4 est.
• Mac Studio M2 Ultra 192GB — ≈30.0 tok/s Q4 est.

All GPUs for Solar Open 2 →

Other models for the T4 16GB

• MiniCPM5-1B — ≈301.1 tok/s Q4 est.
• Qwen3.5-2B — ≈162.6 tok/s Q4 est.
• Llama-3.2-3B-Instruct — ≈101.6 tok/s Q4 est.
• gpt-oss-20b — ≈90.3 tok/s Q4 est.
• gemma-4-26B-A4B-it — ≈85.6 tok/s Q4 est.
• Nanbeige4.2-3B — ≈82.8 tok/s Q4 est.
• Qwen3-4B-Instruct-2507 — ≈81.3 tok/s Q4 est.
• NVIDIA-Nemotron-3-Nano-4B-BF16 — ≈81.3 tok/s Q4 est.

All models on the T4 16GB →