What can Used NVIDIA RTX 4090 24GB run? 281 models fit & the real cost

281 of the 377 models in the Spanvero catalog fit Used NVIDIA RTX 4090 24GB's 24 GB VRAM (at a sensible quant, 16k context). For each: run it locally ($0 compute + electricity), rent an equivalent GPU ($0 markup, as of 2026-07-02), or pay per-token via your own API key (as of 2026-06-29).

Three honest ways to run each model on Used NVIDIA RTX 4090 24GB

• Run it locally: $0 in compute — you pay only electricity (~350 W under load on this card). Local is real money, never a fake "$0".
• Rent an equivalent GPU: from a $0-markup vendor rate (as of 2026-07-02) — you rent on your own account and pay the vendor directly; we never resell compute.
• Skip the box: run the same model through your own API key, paying per million tokens (prices as of 2026-06-29).

What fits Used NVIDIA RTX 4090 24GB (24 GB VRAM)

• 281 of the 377 notable models in the Spanvero catalog fit Used NVIDIA RTX 4090 24GB at a sensible quant (context capped at 16k for the estimate). Most capable first:
• Laguna XS.2 (poolside, 33.4B) — needs ~24 GB at Q4_K_M: run it locally for $0 compute + ~$0.4254/1M in electricity, rent NVIDIA RTX A6000 48GB from $0.49/hr ($0 markup), or ~$0.15/1M via your own API key.
• sarvam 30b (sarvamai, 32.2B) — needs ~22 GB at Q4_K_M: run it locally for $0 compute + ~$0.4131/1M in electricity, rent NVIDIA RTX A6000 48GB from $0.49/hr ($0 markup), or ~$0.36/1M via your own API key (size estimate).
• llm jp 4 32b a3b thinking (llm-jp, 32.1B) — needs ~23 GB at Q4_K_M: run it locally for $0 compute + ~$0.4121/1M in electricity, rent NVIDIA RTX A6000 48GB from $0.49/hr ($0 markup), or ~$0.36/1M via your own API key (size estimate).
• NVIDIA Nemotron 3 Nano 30B A3B BF16 (nvidia, 31.6B) — needs ~22 GB at Q4_K_M: run it locally for $0 compute + ~$0.407/1M in electricity, rent NVIDIA RTX A6000 48GB from $0.49/hr ($0 markup), or ~$0.35/1M via your own API key (size estimate).
• Nemotron Cascade 2 30B A3B (nvidia, 31.6B) — needs ~22 GB at Q4_K_M: run it locally for $0 compute + ~$0.407/1M in electricity, rent NVIDIA RTX A6000 48GB from $0.49/hr ($0 markup), or ~$0.35/1M via your own API key (size estimate).
• Qwen3 30B A3B (Qwen, 30.5B) — needs ~22 GB at Q4_K_M: run it locally for $0 compute + ~$0.3956/1M in electricity, rent NVIDIA RTX A6000 48GB from $0.49/hr ($0 markup), or ~$0.31/1M via your own API key.
• Qwen3 Coder 30B A3B Instruct (Qwen, 30.5B) — needs ~22 GB at Q4_K_M: run it locally for $0 compute + ~$0.3956/1M in electricity, rent NVIDIA RTX A6000 48GB from $0.49/hr ($0 markup), or ~$0.17/1M via your own API key.
• Qwen3 30B A3B Instruct 2507 (Qwen, 30.5B) — needs ~22 GB at Q4_K_M: run it locally for $0 compute + ~$0.3956/1M in electricity, rent NVIDIA RTX A6000 48GB from $0.49/hr ($0 markup), or ~$0.12/1M via your own API key.
• Qwen3 30B A3B Thinking 2507 (Qwen, 30.5B) — needs ~22 GB at Q4_K_M: run it locally for $0 compute + ~$0.3956/1M in electricity, rent NVIDIA RTX A6000 48GB from $0.49/hr ($0 markup), or ~$0.24/1M via your own API key.
• Tongyi DeepResearch 30B A3B (Alibaba-NLP, 30.5B) — needs ~22 GB at Q4_K_M: run it locally for $0 compute + ~$0.3956/1M in electricity, rent NVIDIA RTX A6000 48GB from $0.49/hr ($0 markup), or ~$0.34/1M via your own API key (size estimate).
• lynx instruct 30b (bineric, 30.5B) — needs ~22 GB at Q4_K_M: run it locally for $0 compute + ~$0.3956/1M in electricity, rent NVIDIA RTX A6000 48GB from $0.49/hr ($0 markup), or ~$0.34/1M via your own API key (size estimate).
• North Mini Code 1.0 (CohereLabs, 30.5B) — needs ~22 GB at Q4_K_M: run it locally for $0 compute + ~$0.3956/1M in electricity, rent NVIDIA RTX A6000 48GB from $0.49/hr ($0 markup), or ~$0.34/1M via your own API key (size estimate).
• granite 4.1 30b (ibm-granite, 28.9B) — needs ~24 GB at Q4_K_M: run it locally for $0 compute + ~$0.3789/1M in electricity, rent NVIDIA RTX A6000 48GB from $0.49/hr ($0 markup), or ~$0.33/1M via your own API key (size estimate).
• Gemma 2 27B Instruct (Google, 27B) — needs ~22 GB at Q4_K_M: run it locally for $0 compute + ~$0.3588/1M in electricity, rent NVIDIA RTX A6000 48GB from $0.49/hr ($0 markup), or ~$0.65/1M via your own API key.
• Trinity Mini (arcee-ai, 26.1B) — needs ~19 GB at Q4_K_M: run it locally for $0 compute + ~$0.3492/1M in electricity, rent NVIDIA RTX 3090 24GB from $0.26/hr ($0 markup), or ~$0.10/1M via your own API key.
• LFM2 24B A2B (LiquidAI, 23.8B) — needs ~18 GB at Q4_K_M: run it locally for $0 compute + ~$0.3244/1M in electricity, rent NVIDIA RTX 3090 24GB from $0.26/hr ($0 markup), or ~$0.29/1M via your own API key (size estimate).
• Mistral Small 3 (24B, 2501) (Mistral AI, 23.6B) — needs ~20 GB at Q4_K_M: run it locally for $0 compute + ~$0.3222/1M in electricity, rent NVIDIA RTX 3090 24GB from $0.26/hr ($0 markup), or ~$0.07/1M via your own API key.
• EuroLLM 22B Instruct 2512 (utter-project, 22.6B) — needs ~19 GB at Q4_K_M: run it locally for $0 compute + ~$0.3112/1M in electricity, rent NVIDIA RTX 3090 24GB from $0.26/hr ($0 markup), or ~$0.28/1M via your own API key (size estimate).
• gpt oss safeguard 20b (openai, 21.5B) — needs ~16 GB at Q4_K_M: run it locally for $0 compute + ~$0.2991/1M in electricity, rent NVIDIA RTX 3090 24GB from $0.26/hr ($0 markup), or ~$0.19/1M via your own API key.
• gpt-oss-20b (OpenAI, 21B) — needs ~15 GB at Q4_K_M: run it locally for $0 compute + ~$0.2935/1M in electricity, rent NVIDIA RTX 3090 24GB from $0.26/hr ($0 markup), or ~$0.08/1M via your own API key.

• Laguna XS.2 — full cost breakdown
• sarvam 30b — full cost breakdown
• llm jp 4 32b a3b thinking — full cost breakdown
• NVIDIA Nemotron 3 Nano 30B A3B BF16 — full cost breakdown
• Nemotron Cascade 2 30B A3B — full cost breakdown
• Qwen3 30B A3B — full cost breakdown
• Qwen3 Coder 30B A3B Instruct — full cost breakdown
• Qwen3 30B A3B Instruct 2507 — full cost breakdown
• Qwen3 30B A3B Thinking 2507 — full cost breakdown
• Tongyi DeepResearch 30B A3B — full cost breakdown
• lynx instruct 30b — full cost breakdown
• North Mini Code 1.0 — full cost breakdown
• granite 4.1 30b — full cost breakdown
• Gemma 2 27B Instruct — full cost breakdown
• Trinity Mini — full cost breakdown
• LFM2 24B A2B — full cost breakdown
• Mistral Small 3 (24B, 2501) — full cost breakdown
• EuroLLM 22B Instruct 2512 — full cost breakdown
• gpt oss safeguard 20b — full cost breakdown
• gpt-oss-20b — full cost breakdown

The honest cost of owning Used NVIDIA RTX 4090 24GB

• Street price $2,270.00 (as of 2026-07-02; BestValueGPU eBay used-price tracker, Jun 2026 (~$2,268)) — amortized over 3 years that's ~$2.0731/day whether or not you're generating.
• Electricity: ~350 W under sustained inference at $0.1883/kWh (EIA Electric Power Monthly Table 5.6.A — U.S. residential average, Apr 2026, as of 2026-07-02) — the per-1M-token figures above already include this at each model's speed.
• Straight talk: for the small models a 24 GB VRAM box runs, hosted APIs are often cheaper per token. Own local for privacy, offline use, and unlimited runs — not to save money on tokens.

Too big for Used NVIDIA RTX 4090 24GB — rent or use an API instead

• These need more than the 24 GB VRAM on this card. Closest first — you can still run them on a rented GPU ($0 markup) or via your own API key:
• Yi-1.5-34B-Chat (01.AI, 34.4B) — needs ~25 GB; rent NVIDIA RTX A6000 48GB from $0.49/hr, or ~$0.38/1M via your own API key (size estimate).
• Qwen3-32B (Alibaba, 32.8B) — needs ~25 GB; rent NVIDIA RTX A6000 48GB from $0.49/hr, or ~$0.18/1M via your own API key.
• granite 4.0 h small (ibm-granite, 32.2B) — needs ~25 GB; rent NVIDIA RTX A6000 48GB from $0.49/hr, or ~$0.36/1M via your own API key (size estimate).
• Qwen2.5-Coder 32B Instruct (Alibaba, 32B) — needs ~25 GB; rent NVIDIA RTX A6000 48GB from $0.49/hr, or ~$0.83/1M via your own API key.
• Gemma 4 31B IT NVFP4 (nvidia, 20.9B) — needs ~25 GB; rent NVIDIA RTX A6000 48GB from $0.49/hr, or ~$0.27/1M via your own API key (size estimate).
• Seed OSS 36B Instruct (ByteDance-Seed, 36.2B) — needs ~27 GB; rent NVIDIA RTX A6000 48GB from $0.49/hr, or ~$0.39/1M via your own API key (size estimate).

Keep exploring

• Check a different GPU or Mac
• All GPUs — rental rates compared
• Free cost calculator — local vs rent vs API
• All models — the real cost to run each
• How we stay honest

Open the free Spanvero advisor → · GPU rates as of 2026-07-02, API prices as of 2026-06-29. Honest, $0-markup — your own accounts, we never resell compute. © 2026 Cynosure LLC.