← AI Hashrate 中文

Can RTX 4060 8GB run OLMo-2-1124-13B-Instruct?

NVIDIA · 8 GB GDDR6 · 272 GB/s bandwidth · allenai · 13.7B · model ctx up to NoneK

FF · No FitComposite score: VRAM headroom × 0.5 + tok/s speed × 0.5. Q4 @ 8K context.
CUDA

No — OLMo-2-1124-13B-Instruct at Q4 needs ≈ 9.91 GB even at 4K context, beyond the RTX 4060 8GB's 8 GB GDDR6. It would only run with heavy CPU/disk offload. Decode at Q4/8K: ≈ 18.0 tok/s (estimated, batch 1). FP16 needs ≈ 31.14 GB — does not fit on this card. Estimates come from memory-bandwidth math; rows tagged measured override estimates. Relative ranking is more reliable than absolute tok/s. Methodology.

Cheapest GPU that can: H100 80GB SXM5 See fit details →or☁️ Rent on RunPod from ~$0.50/hr

Fit & speed by quant and context

QuantContextVRAM neededFits?tok/s (decode)
Q44K9.91 GBNo23.4 ± est.
Q48K default11.28 GBNo18.0 ± est.
Q432K19.5 GBNo6.0 ± est.
FP164K29.77 GBNo0.7 ± est.
FP168K default31.14 GBNo0.7 ± est.
FP1632K39.36 GBNo

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

QuantWeightsKV cacheOverheadTotal neededRTX 4060 8GB VRAM
Q47.54 GB2.74 GB1 GB11.28 GB8 GB
FP1627.4 GB2.74 GB1 GB31.14 GB8 GB

Other GPUs that run OLMo-2-1124-13B-Instruct

All GPUs for OLMo-2-1124-13B-Instruct →

Other models for the RTX 4060 8GB

All models on the RTX 4060 8GB →