RUNLOCALAIv38
→WILL IT RUNBEST GPUCOMPARETROUBLESHOOTSTARTPULSEMODELSHARDWARETOOLSBENCH
RUNLOCALAI

Independently operated catalog for local-AI hardware and software. Hand-written verdicts. Source-cited claims. Reproducible commands when we have them.

OP·Fredoline Eruo
DIR
  • Models
  • Hardware
  • Tools
  • Benchmarks
TOOLS
  • Will it run?
  • Compare hardware
  • Cost vs cloud
  • Choose my GPU
  • Quick answers
REF
  • All buyer guides
  • Methodology
  • Glossary
  • Errors KB
  • Trust
EDITOR
  • About
  • Author
  • How we make money
  • Editorial policy
  • Contact
LEGAL
  • Privacy
  • Terms
  • Sitemap
MAIL · MONTHLY DIGEST
Get monthly local AI changes
Monthly recap. No spam.
DISCLOSURE

Some links on this site are affiliate links (Amazon Associates and other first-class retailers). When you buy through them, we earn a small commission at no extra cost to you. Affiliate links do not influence our verdicts — there are cards we rate highly that we don't have affiliate relationships with, and cards that sell well that we refuse to recommend. Read more →

© 2026 runlocalai.coIndependently operated
RUNLOCALAI · v38
Will it run? / NVIDIA RTX 4090 48GB (China-mod) / coding

What can NVIDIA RTX 4090 48GB (China-mod) run for coding?

Build: NVIDIA RTX 4090 48GB (China-mod) + — + 32 GB RAM (windows)

Memory: 48 GB VRAM + 32 GB system RAM
Runner: llama.cpp / Ollama (CUDA)
AnyChatCodingAgentsReasoningVisionLong contextCreative

Runs comfortably
107 models

Ranked by fit for coding use case + predicted speed. Click a row for VRAM breakdown.

#1CodeGemma 7B
7B
gemma
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 17.9 GBHeadroom: 30.1 GBTTFT: fast
ollama run codegemma:7b
155
tok/s
E
Weights
4.23 GB
KV cache
3.50 GB
Activations
8.40 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~217 ms (fast)
Model details →Run-on benchmark page →
#2DeepSeek Coder V2 Lite (16B)
16B
deepseek
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 28.1 GBHeadroom: 19.9 GBTTFT: fast
ollama run deepseek-coder-v2:16b
68
tok/s
E
Weights
9.66 GB
KV cache
8.00 GB
Activations
8.68 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~496 ms (fast)
Model details →Run-on benchmark page →
#3Qwen 2.5 7B Instruct
7B
qwen
Commercial OK
Quant: Q8_0Context: 8,192VRAM: 18.3 GBHeadroom: 29.7 GBTTFT: fast
ollama run qwen2.5:7b
88
tok/s
E
Weights
7.44 GB
KV cache
0.47 GB
Activations
8.56 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~217 ms (fast)
Model details →Run-on benchmark page →
#4Qwen 3 8B
8B
qwen
Commercial OK
Quant: Q8_0Context: 8,192VRAM: 22.9 GBHeadroom: 25.1 GBTTFT: fast
ollama run qwen3:8b
77
tok/s
E
Weights
8.50 GB
KV cache
4.00 GB
Activations
8.62 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~248 ms (fast)
Model details →Run-on benchmark page →
#5Mistral Small 3 24B
24B
mistral
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 37.2 GBHeadroom: 10.8 GBTTFT: noticeable
ollama run mistral-small:24b
45
tok/s
E
Weights
14.49 GB
KV cache
12.00 GB
Activations
8.92 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~744 ms (noticeable)
Model details →Run-on benchmark page →
#6Qwen 3 14B
14B
qwen
Commercial OK
Quant: Q8_0Context: 8,192VRAM: 32.6 GBHeadroom: 15.4 GBTTFT: fast
ollama run qwen3:14b
44
tok/s
E
Weights
14.88 GB
KV cache
7.00 GB
Activations
8.94 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~434 ms (fast)
Model details →Run-on benchmark page →
#7Qwen 2.5 14B Instruct
14B
qwen
Commercial OK
Quant: Q8_0Context: 8,192VRAM: 32.6 GBHeadroom: 15.4 GBTTFT: fast
ollama run qwen2.5:14b
44
tok/s
E
Weights
14.88 GB
KV cache
7.00 GB
Activations
8.94 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~434 ms (fast)
Model details →Run-on benchmark page →
#8Llama 3.1 8B Instruct
8B
llama
Commercial OK
Quant: FP16Context: 8,192VRAM: 27.9 GBHeadroom: 20.1 GBTTFT: fast
ollama run llama3.1:8b
41
tok/s
E
Weights
16.00 GB
KV cache
1.07 GB
Activations
8.99 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~248 ms (fast)
Model details →Run-on benchmark page →
#9StarCoder 2 3B
3B
other
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 13.4 GBHeadroom: 34.6 GBTTFT: instant
362
tok/s
E
Weights
1.81 GB
KV cache
1.50 GB
Activations
8.28 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~93 ms (instant)
Model details →Run-on benchmark page →
#10StarCoder 2 7B
7B
other
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 17.9 GBHeadroom: 30.1 GBTTFT: fast
155
tok/s
E
Weights
4.23 GB
KV cache
3.50 GB
Activations
8.40 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~217 ms (fast)
Model details →Run-on benchmark page →
#11Codestral Mamba 7B
7B
mistral
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 17.9 GBHeadroom: 30.1 GBTTFT: fast
155
tok/s
E
Weights
4.23 GB
KV cache
3.50 GB
Activations
8.40 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~217 ms (fast)
Model details →Run-on benchmark page →
#12Qwen 2.5 Coder 3B
3B
qwen
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 13.4 GBHeadroom: 34.6 GBTTFT: instant
362
tok/s
E
Weights
1.81 GB
KV cache
1.50 GB
Activations
8.28 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~93 ms (instant)
Model details →Run-on benchmark page →

Runs with tradeoffs
22 models

Tight VRAM, partial CPU offload, or context-limited.

Qwen 3 30B-A3B
30B
qwen
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 44.0 GBHeadroom: 4.0 GBTTFT: noticeable
  • • Tight VRAM fit — only 4.0 GB headroom left for context growth
ollama run qwen3:30b
36
tok/s
E
Weights
18.11 GB
KV cache
15.00 GB
Activations
9.10 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~930 ms (noticeable)
Model details →Run-on benchmark page →
Gemma 4 31B Dense
31B
gemma
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 45.1 GBHeadroom: 2.9 GBTTFT: noticeable
  • • Tight VRAM fit — only 2.9 GB headroom left for context growth
ollama run gemma4:31b
35
tok/s
E
Weights
18.72 GB
KV cache
15.50 GB
Activations
9.13 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~961 ms (noticeable)
Model details →Run-on benchmark page →
Qwen 3 32B
32B
qwen
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 46.3 GBHeadroom: 1.7 GBTTFT: noticeable
  • • Tight VRAM fit — only 1.7 GB headroom left for context growth
ollama run qwen3:32b
34
tok/s
E
Weights
19.32 GB
KV cache
16.00 GB
Activations
9.16 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~992 ms (noticeable)
Model details →Run-on benchmark page →
Qwen 2.5 32B Instruct
32B
qwen
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 46.3 GBHeadroom: 1.7 GBTTFT: noticeable
  • • Tight VRAM fit — only 1.7 GB headroom left for context growth
ollama run qwen2.5:32b
34
tok/s
E
Weights
19.32 GB
KV cache
16.00 GB
Activations
9.16 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~992 ms (noticeable)
Model details →Run-on benchmark page →
Codestral 22B
22B
mistral
Quant: Q8_0Context: 8,192VRAM: 45.5 GBHeadroom: 2.5 GBTTFT: noticeable
  • • Tight VRAM fit — only 2.5 GB headroom left for context growth
ollama run codestral:22b
28
tok/s
E
Weights
23.38 GB
KV cache
11.00 GB
Activations
9.36 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~682 ms (noticeable)
Model details →Run-on benchmark page →
Qwen 2.5 Coder 32B Instruct
32B
qwen
Commercial OK
Quant: Q8_0Context: 8,192VRAM: 47.8 GBHeadroom: 0.2 GBTTFT: noticeable
  • • Tight VRAM fit — only 0.2 GB headroom left for context growth
ollama run qwen2.5-coder:32b
19
tok/s
E
Weights
34.00 GB
KV cache
2.15 GB
Activations
9.89 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~992 ms (noticeable)
Model details →Run-on benchmark page →
Llama 3.1 70B Instruct
70B
llama
Commercial OK
Quant: Q4_K_MContext: 2,048VRAM: 57.0 GBHeadroom: 10.2 GBTTFT: slow
  • • Partial CPU offload: ~16% of layers run on CPU
ollama run llama3.1:70b
16
tok/s
E
Weights
42.26 GB
KV cache
8.75 GB
Activations
4.16 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~2169 ms (slow)
Model details →Run-on benchmark page →
Llama 3.3 70B Instruct
70B
llama
Commercial OK
Quant: Q5_K_MContext: 8,192VRAM: 63.2 GBHeadroom: 4.0 GBTTFT: slow
  • • Partial CPU offload: ~24% of layers run on CPU
ollama run llama3.3:70b
14
tok/s
E
Weights
48.13 GB
KV cache
2.68 GB
Activations
10.60 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~2169 ms (slow)
Model details →Run-on benchmark page →

What if you upgraded?

Hypothetical scenarios. We re-ran the compatibility engine for each.

+32 GB system RAM

~$80–150

Doubles your CPU-offload working set. Helps when models don't quite fit in VRAM.

Unlocks: 17 new comfortable, 28 new tradeoff

  • • Gemma 3 1B
  • • Llama 3.2 1B Instruct
  • • Gemma 4 E2B (Effective 2B)
  • • Whisper Large v3
Shop this upgrade↗

Upgrade to NVIDIA H100 SXM

see current pricing

80 GB VRAM (vs your 48 GB) plus a bandwidth jump from ~1008 GB/s to ~3350 GB/s.

Unlocks: 40 new comfortable

  • • Gemma 3 1B
  • • Llama 3.2 1B Instruct
  • • Gemma 4 E2B (Effective 2B)
  • • QwQ 32B Preview
Shop this upgrade↗

Add a second NVIDIA RTX 4090 48GB (China-mod)

~$2400

Tensor parallelism splits the model across both cards, effectively doubling VRAM. Bandwidth doesn't double — runs ~1.5× the single-card speed in practice.

Unlocks: 46 new comfortable

  • • Gemma 3 1B
  • • Llama 3.2 1B Instruct
  • • Gemma 4 E2B (Effective 2B)
  • • QwQ 32B Preview
Shop this upgrade↗

Some links above are affiliate links. We may earn a commission at no extra cost to you. How we make money.

Won't run
top 5 popular models

Need more memory than you have. Shown for orientation.

DeepSeek V4 Pro (1.6T MoE)
1600B
deepseek
Commercial OK

Even with CPU offload, needs more memory than your VRAM (48 GB) + 60% of system RAM (19 GB) combined.

—
Qwen 3.5 235B-A17B (MoE)
397B
qwen
Commercial OK

Even with CPU offload, needs more memory than your VRAM (48 GB) + 60% of system RAM (19 GB) combined.

—
Qwen 3 235B-A22B
235B
qwen
Commercial OK

Even with CPU offload, needs more memory than your VRAM (48 GB) + 60% of system RAM (19 GB) combined.

—
Llama 4 Scout
109B
llama
Commercial OK

Even with CPU offload, needs more memory than your VRAM (48 GB) + 60% of system RAM (19 GB) combined.

—
DeepSeek R1 (671B reasoning)
671B
deepseek
Commercial OK

Even with CPU offload, needs more memory than your VRAM (48 GB) + 60% of system RAM (19 GB) combined.

—

How to read these numbers

M
Measured — we ran this exact combo on owner hardware.

~
Extrapolated — predicted from a measured benchmark on similar-bandwidth hardware.

E
Estimated — pure formula based on VRAM bandwidth and model architecture.

Full methodology →

Want a specific benchmark we don't have? Email support@runlocalai.co and we'll prioritize it.