Ranking updated on 09/10/2026
The RTX 4080 Super (16 GB GDDR6X, 736 GB/s) is the boosted Ada Lovelace tier 1 variant. ~5% faster than the 4080. 16 GB unlocks 24B models in Q4_K_M at 30–35 tok/s.
RTX 4080 Super : purchasing alternative available for local AI — RTX 5080 16 GB :
A mini PC is a complete machine: check the required memory and software compatibility. It does not replace macOS/MLX or CUDA.
Which PC should you choose for your budget? Our picks from €800 to €3,500 →
Affiliate links — QuelLLM may earn a commission on purchases at no extra cost to you, which does not influence the ranking (established independently). As an Amazon Associate, BestLLMfor earns from qualifying purchases.
Dense 14B with hybrid thinking. Equals Qwen 2.5 32B Based on STEM/code.
ollama run qwen3:14b
MIT 14B reasoner. Beats R1-Distill-Llama-70B on AIME/GPQA with 50× fewer parameters.
ollama run phi4-reasoning:14b
Exceptional reasoning for its size. STEM-focused.
ollama run phi4:14b
Coding 14B. HumanEval 89.6, LiveCodeBench 37.1. VRAM sweet spot for self-hosted coding.
ollama run qwen2.5-coder:14b
Distilled R1 Qwen 14B. AIME24 69.7, MATH-500 93.9. Outperforms o1-mini on many benchmarks.
ollama run deepseek-r1:14b
Dense 14B Apache 2.0. MMLU 79.7, HumanEval 83.5. 29+ languages. Good compromise.
ollama run qwen2.5:14b
24B coding specialist, Apache 2.0. 72.2% SWE-Bench. 256k ctx, FR lab.
ollama run devstral-small2:24b
| Rank | Model | Params | Q4 VRAM | Context | License | On RTX 4080 Super |
|---|---|---|---|---|---|---|
| #1 | Qwen 3 14B | 14B | 9 GB | 131 072 | Apache 2.0 | 55 tok/s · Q8 |
| #2 | Phi-4 Reasoning 14B | 14B | 9 GB | 32 768 | MIT | 55 tok/s · Q8 |
| #3 | Phi-4 14B | 14B | 9 GB | 16 384 | MIT | 55 tok/s · Q8 |
| #4 | Qwen 2.5 Coder 14B Instruct | 14B | 9 GB | 131 072 | Apache 2.0 | 55 tok/s · Q8 |
| #5 | DeepSeek R1 Distill Qwen 14B | 14B | 9 GB | 131 072 | MIT | 55 tok/s · Q8 |
| #6 | Qwen 2.5 14B Instruct | 14B | 9 GB | 131 072 | Apache 2.0 | 55 tok/s · Q8 |
| #7 | Devstral Small 2 24B | 24B | 14 GB | 256 000 | Apache 2.0 | 40 tok/s · Q4_K_M |
Your private, free ChatGPT on your machine in 1 hour — LM Studio, Ollama, Open WebUI, your documents, no cloud.
Free memo
Get the memo VRAM → best coding model → Ollama command (one screen, copy and paste). Then switch to the Copilote Local kit for a setup that actually works.
The Local Copilot kit — the Ollama + Cline + Aider configs are ready to paste, with tuned Modelfiles, troubleshooting, and lifetime online access →No spam. Unsubscribe in 1 click. Your data stays with us (never resold).
Your card → the best coding model to run locally, and the exact Ollama command:
| Your VRAM | Typical GPUs / Macs | Recommended coding model | Command Ollama |
|---|---|---|---|
| 8 GB | RTX 4060 / 3060 · M1-M2 16 GB | Qwen 3.5 9B (Q4, 6.6 GB — 256k context) | ollama run qwen3.5:9b |
| 12 GB | RTX 3060 12 GB / 4070 / 5070 | Qwen 3.5 9B (Q8, 11 GB) or Gemma 4 12B (7.6 GB) | ollama run qwen3.5:9b-q8_0 |
| 16 GB | RTX 5070 Ti / 4080 / 5080 · RX 9070 XT · M4 24 GB | Devstral 24B (Q4, 14 GB) — coding-agent specialist | ollama run devstral:24b |
| 24 GB | RTX 3090 / 4090 · RX 7900 XTX · M4 Pro 48 GB | Qwen 3.8 27B (Q4, 18 GB) — the “close to Copilot” option | ollama run qwen3.8:27b |
| 32 GB | RTX 5090 | Qwen 3.6 35B-A3B (Q4, 23 GB) — fast MoE | ollama run qwen3.6:35b |
| 48 GB+ | Mac M4 Max 64 GB · M2 Ultra 128 GB | Qwen3-Coder 30B-A3B (Q8, 32 GB — 256k context) | ollama run qwen3-coder:30b-a3b-q8_0 |
-base : ollama run qwen2.5-coder:7b-base — it’s still the reference for this specific use case. ⚠️ Qwen 3.8: its reasoning is set very high by default and it “overthinks” simple requests — lower it to low (or turn it off) on first launch. ⚠️ License trap: Codestral 22B = Mistral Non-Production License → prohibited for coding at work. Qwen 3.5/3.8, Gemma 4, and Devstral are Apache 2.0. 💡 Running out of memory? Keep ~1.5 GB of VRAM free for context, or drop down one quantization level.🔌 To connect it to VS Code: Cline (multi-file agent), Aider (CLI) or Tabby/Twinny (FIM autocomplete) — they all connect to Ollama locally. The kit Local Copilot — ready-to-paste configs + tested setup — is available: /copilote-local.
Filter: Q4_K_M ≤ 14 GB. Bonus: 7–14B and 13–24B. Boost variant ~5% vs. standard 4080.
Criteria considered:
The scoring is fully transparent: see our methodology for details on VRAM/tokens/sec calculations.
RTX 4080 Super vs 4080?
Even 16 GB. 4080 Super = +5% performance (~7% CUDA cores, slightly faster GDDR6X). Imperceptible difference for most LLMs. Prefer the cheaper used option. See RTX 4080.
4080 Super vs. 5080?
5080 = ~25% faster (GDDR7 vs. GDDR6X) but ~€600 more when new. Used 4080 Super at ~€700 = better value. See RTX 5080.
Llama 70B on a 4080 Super?
No — Q4 (~40 GB) does not fit in 16 GB. Q2_K (~28 GB) does not fit either. For 70B locally on consumer hardware, target RTX 4090/5090 24+ GB or Mac Studio 64+ GB.
LLM sweet spot on a 4080 Super?
Mistral Small 24B Q4 (~13 GB) at 30 tok/s or Qwen 3 14B Q6 (~12 GB) at 50 tok/s. For code, Qwen 2.5 Coder 14B Q5. See code ranking.
Learn more with our detailed head-to-head matchups of the finalists: