NVIDIA GeForce RTX 4070

The default 13B card: 504 GB/s on a 192-bit bus with 12 GB — full-quality 13B with a real context, at 200 W.

12 GB GDDR6X 504 GB/s 192-bit bus 200 W TDP

Technical specifications

GPU die
AD103
CUDA cores
5888
Boost clock
~2.48 GHz
VRAM
12 GB GDDR6X
Memory bus
192-bit
Memory bandwidth
504 GB/s
PCIe
4.0 x16
NVLink
none
TDP
200 W
Release
Feb 2023

What these specs mean for llama.cpp

The bus the tier deserves

The jump from the 4060 Ti's 128-bit to the 4070's 192-bit is where Ada starts earning its keep: 504 GB/s is ~75% more than 288, and token generation scales with it — ~55–72 t/s on 8B, ~35–46 t/s on 13B Q4. Prompt processing (600–800 t/s on 13B) is the same story. This is the first 40-series card where the speed and the VRAM stop pulling in opposite directions.

12 GB, honestly assessed

12 GB is the 13B tier's home: Q4_K_M at 32K+ with room, Q5_K_M at 8K, 8B at Q8 with a big context. It is not the 27B tier's — 19 GB of Q4 weights is 7 GB past the lid — and not the MoE 30B tier's at full quant either. The ceiling is real; the card inside it is a solid mid-range.

200 W of mid-range

Between the 4060's 115 W and the 4070 Ti's 285 W, the 4070 is the efficiency midpoint of the series — a fine 24/7 card with an easy 650 W PSU requirement.

The short form 13B at full quality, 32K context, no compromises — that is the 4070's job and it does it well. 27B ambitions belong to the 16 GB tier (4070 Ti Super) or a second card.

The MTP question: if your model ships a built-in multi-token-prediction head (Qwen 3.5/3.6/3.8, Gemma 4), the one-flag speculative pass (--spec-type draft-mtp --spec-draft-n-max 2 --parallel 1) is worth +40–65% (estimated) on token generation here; no 4070 A/B in the record; mid-bandwidth Ada cards land in the middle of the measured spread at n-max 2. The price is ~0.6–2 GB of VRAM, the output is bit-identical to the unaccelerated run, and models without MTP heads (Llama class) see nothing. Full tuning rules in the MTP guide.

Expected performance (Q4_K_M, single GPU)

ModelSize on GPUToken genToken gen (MTP est.)Prompt proc.Fits?
8B Q8_0~9.7 GB~55–72 t/s~77–119 t/s~600–800 t/s✅ 32K ctx
13B Q4_K_M~8.5 GB~35–46 t/s~49–76 t/s~650–850 t/s✅ 32K+ ctx
13B Q5_K_M~11.9 GB~25–33 t/s~35–54 t/s~550–700 t/s⚠️ 8K ctx
27B Q4_K_M~19 GB———❌ ~7 GB over

MTP est. = rough prior with --spec-type draft-mtp for models that ship MTP heads (Qwen 3.5/3.6/3.8, Gemma 4); Llama-class models get no MTP speedup. Actual gains depend on model, quant, context length, llama.cpp build and your card — sweep --spec-draft-n-max, don't trust the column blindly. MTP guide →

Figures assume full offload (-ngl 99), ~8K context and batch 2048. Token speed tracks the memory bus almost linearly, so treat ranges as class estimates — your llama.cpp build and the model's attention layout shift them.

27B models (Qwen class) — Q4 / Q5 / Q6 / Q8

QuantGGUF size1× 4070 (12 GB)2× 4070 (24 GB)
Q4_K_M~19 GB❌ ~7 GB over⚠️ ~16–20 t/s, 8K ctx
Q5_K_M~22 GB❌❌ ~2 GB over
Q6_K~25 GB❌❌
Q8_0~33 GB❌❌

Single-card, 27B is a hard no. A pair (24 GB) gets Q4_K_M on at 8K context — workable, unhurried. Three cards (36 GB) is the proper 4070 27B build: Q5 at 8–16K around ~14–18 t/s.

Practical tips

Multi-card: 2×, 3×, 4×

PCIe 4.0, ~25–35% tax, 12 GB tiles: 2× reaches the 27B Q4 line at 24 GB, 3× reaches Q5 at 36 GB.

BuildGPU powerVRAM13B Q827B Q4_K_M27B Q5_K_M
2× 4070400 W24 GB~40–52 t/s⚠️ ~16–20 t/s, 8K ctx❌ ~2 GB over
3× 4070600 W36 GB(overkill)~19–24 t/s, 16K+ ctx~14–18 t/s, 8–16K ctx

The 3-way build is where 27B Q5 with a real context becomes a 4070 story. It wants a board with three working x8/x16 lanes — common in X670E/WRX80, uncommon elsewhere.

The price check 2× 4070 Ti Super (570 W, 32 GB) runs the same 27B Q4/Q5 work faster with fewer cards. The 4070 builds win on budget; the Ti Super pair wins on everything else.

Offloading (CPU RAM)

When a model won't fit, -ngl N parks the tail layers on the CPU. The mental model that saves pain: offloaded layers don't run at a fraction of GPU speed — they run at CPU speed. Handy moves: peel the last 2–4 layers off to reclaim ~0.5–1 GB of VRAM for almost nothing; keep MoE experts on CPU with --n-cpu-moe; park the output embedding with -ot output=CPU on big models. Budget 32 GB system RAM for 27B-class offload, 64 GB for 70B. And never offload the KV cache — the session becomes unusable.

3-card example llama-server -m 27b-q5_k_m.gguf -ngl 99 -sm layer -c 16384 --tensor-split 1,1,1

Recommended llama.cpp command

llama-server -m llama-3.1-13b-q4_k_m.gguf -ngl 99 -c 32768 -b 2048 -ub 2048 --host 127.0.0.1 --port 8080

Verdict

The 4070 is the safe answer to 'which 40-series card runs 13B well': 504 GB/s, 12 GB, 200 W, and a 32K context at full quality without a knob turned. It has no drama and no ceiling surprise until 27B arrives — at which point it's a multi-card project. For the 13B lane, it's the series' default for a reason.

← RTX 4060 Ti 16 GB next: RTX 4070 Super →