01
Gemma 4 26B-A4B
Gemma / 26B / Q4_K_M / ~16 GB
Best for: Chat, Coding, Multimodal·Pop: 86/100
Perf: ~85 tok/s · first token ~0.4s
Local OKOK
Fits in 24 GB VRAM with room to spare. Best for chat, coding, multimodal on RTX 4090.
Best Local AI Models for RTX 4090 (24GB)
The RTX 4090 is the current king of local AI inference. With 24GB GDDR6X it handles everything from small chat models (~104 tokens per second on 8B) to 32B parameter reasoning models (an estimated 32 tok/s) with ease. The gold standard for serious AI enthusiasts.
Quick answer
The best local LLM for the RTX 4090 is Gemma 4 26B-A4B at ~85 tok/s on its 24GB VRAM. It uses ~16GB of VRAM; the RTX 4090 handles up to 32B parameter models at Q4. A 32B model runs at ~32 tok/s.
$ollama run gemma4:26b
TOP PICK
Gemma 4 26B-A4B
EST. SPEED
~85 tok/s
VRAM NEEDED
~16 GB
Speeds are ModelFit estimates from memory bandwidth and model size, not measured benchmarks.
*Launch MSRP was $1,599; the card is now discontinued and current market pricing runs well above that
RTX 4090 Estimated Tokens/sec by Model Size
Q4_K_M · ModelFit estimate| Model Size | Est. Speed | Fit on 24GB |
|---|---|---|
| 7B | ~117 tok/s | Fits in VRAM |
| 14B | ~65 tok/s | Fits in VRAM |
| 20B MoE (3.6B active) | ~99 tok/s | Fits in VRAM |
| 32B | ~32 tok/s | Fits in VRAM |
| 35B MoE (3B active) | ~84 tok/s | Fits in VRAM |
| 70B | ~2 tok/s | CPU offload (slow) |
| 120B MoE (5.1B active) | ~14 tok/s | CPU offload (slow) |
ModelFit estimates, not measured benchmarks: anchored to an 8B-class Q4_K_M model at 16K context on the RTX 4090's 1008 GB/s bandwidth, then scaled by model size. MoE rows scale by active parameters (decode reads only the active experts), so a 35B MoE runs far faster than a dense 32B. "CPU offload" sizes exceed the 24GB VRAM; dense models slow to a crawl there, MoE models degrade less because hot experts stay GPU-resident.
Context costs VRAM too. Gemma 4 26B-A4B loads ~16 GB of weights; at 16k context the KV cache adds ~4.0 GB (still fits the ~22 GB usable VRAM), and at 64k it adds ~16.0 GB (exceeds the budget, use a smaller quant or a q8_0 KV cache).
KV-cache figures assume an fp16 cache, the llama.cpp/Ollama default. Standard GQA models use a size-class estimate (8 KV heads x 128 head dim class); hybrid linear-attention models (Qwen3.5/3.6, Qwen3-Next) use the exact per-token cost from their published config, since only their sparse full-attention layers cache KV. A q8_0 KV cache roughly halves either figure. Estimates, not measurements.
Where to Buy the RTX 4090
≈ $2,574 street · Launch MSRP was $1,599; the card is now discontinued and current market pricing runs well above thatStorage & accessories for your model library
Internal NVMe SSD · 2TB~$170
A Gen4 M.2 drive keeps your whole GGUF and quant collection on fast local storage, loading models straight off NVMe.
Check price on AmazonUSB4 NVMe Enclosure~$80
40Gbps external storage fast enough to run models from. Pair it with an M.2 drive for a portable model vault.
Check price on AmazonModelFit may earn a commission on purchases through these links, at no extra cost to you. Prices shown are approximate street references.
RTX 4090 VRAM for AI: What Actually Fits?
24GB GDDR6X at 1,008 GB/s gives the RTX 4090 enormous headroom. 32B models at Q4 (~20GB) load fully with 3GB left for KV cache. 14B models at Q5 or Q6 fit easily for higher quality inference. At 104 tok/s with 8B models, the 4090 delivers near-instant responses. The only consumer card faster is the RTX 5090 (145 tok/s, 32GB). For 24GB workloads, the 4090 remains unmatched in speed.
Recommended Models
registry-verified10 models02
Qwen3.5 27B Instruct
Qwen / 27B / Q4_K_M / ~16 GB
Best for: Chat, Coding, Complex reasoning·Pop: 82/100
Perf: ~37 tok/s · first token ~0.4s
Local OKOK
Fits in 24 GB VRAM with room to spare. Best for chat, coding, complex reasoning on RTX 4090.
03
GPT-OSS 20B
GPT-OSS / 21B / MXFP4 / ~13.8 GB
Best for: Chat, Coding, Reasoning·Pop: 85/100
Perf: ~87 tok/s · first token ~0.4s
Local OKExcellent
Fits in 24 GB VRAM with room to spare. Best for chat, coding, reasoning on RTX 4090.
04
Qwen3.6 27B
Qwen / 27B / Q4_K_M / ~18 GB
Best for: Coding, Quality, Long context·Pop: 92/100
Perf: ~37 tok/s · first token ~0.4s
Local OKOK
Fits in 24 GB VRAM with room to spare. Best for coding, quality, long context on RTX 4090.
05
LFM2 24B-A2B Instruct
LFM2 / 24B / Q4_K_M / ~14 GB
Best for: Local AI agents, privacy-first tool calling, MCP workflows·Pop: 80/100
Perf: ~118 tok/s · first token ~0.3s
Local OKExcellent
Fits in 24 GB VRAM with room to spare. Best for local ai agents, privacy-first tool calling, mcp workflows on RTX 4090.
06
Gemma 4 12B (Q8)
Gemma / 12B / Q8_0 / ~12.8 GB
Best for: Chat, Coding, Multimodal·Pop: 80/100
Perf: ~46 tok/s · first token ~0.4s
Local OKExcellent
Fits in 24 GB VRAM with room to spare. Best for chat, coding, multimodal on RTX 4090.
07
Qwen3.5 35B-A3B Instruct
Qwen / 35B / Q4_K_M / ~20 GB
Best for: Reasoning, Coding, Agent scenarios·Pop: 90/100
Perf: ~84 tok/s · first token ~1.0s
Local OKOK
Fits in 24 GB VRAM with room to spare. Best for reasoning, coding, agent scenarios on RTX 4090.
08
Qwen3 14B
Qwen / 14B / Q4_K_M / ~11 GB
Best for: Coding, Quality·Pop: 84/100
Perf: ~65 tok/s · first token ~0.4s
Local OKExcellent
Fits in 24 GB VRAM with room to spare. Best for coding, quality on RTX 4090.
09
Qwen3 14B (Q8)
Qwen / 14B / Q8_0 / ~15.9 GB
Best for: Coding, Quality·Pop: 84/100
Perf: ~40 tok/s · first token ~0.4s
Local OKOK
Fits in 24 GB VRAM with room to spare. Best for coding, quality on RTX 4090.
10
Qwen3.5 9B Instruct
Qwen / 9B / Q4_K_M / ~7 GB
Best for: Quality, Coding, Reasoning·Pop: 86/100
Perf: ~94 tok/s · first token ~0.4s
Local OKExcellent
Fits in 24 GB VRAM with room to spare. Best for quality, coding, reasoning on RTX 4090.
Models Too Big for 24GB? Rent a Cloud GPU
by the hourThe RTX 4090 tops out around up to 32b parameter models. For anything bigger, an hourly rented GPU runs the same open weights with the same Ollama workflow, billed by the hour, no hardware purchase needed.
RunPodHourly GPU pods (RTX 4090 to H100) with one-click Ollama/vLLM templates.Rent
Vast.aiMarketplace of rented GPUs, usually the cheapest per-hour prices.Rent
ModelFit may earn a commission on sign-ups made through these links, at no extra cost to you.
Similar GPUs for Local AI
Compatible Model Families
Qwen
Alibaba Cloud: Widest size range (0.5B to 235B)
LlamaMeta: Most popular open-weight model family
DeepSeekDeepSeek AI: Best-in-class reasoning with R1 models
MistralMistral AI: Excellent performance-per-parameter ratio
GemmaGoogle DeepMind: Excellent quality at small sizes (1B-9B)
PhiMicrosoft: Best quality-per-gigabyte at small sizes
RTX 4090 FAQ: Common Questions
How much VRAM does the RTX 4090 have for LLMs?
The RTX 4090 has 24GB GDDR6X VRAM with 1,008 GB/s bandwidth. About 23GB is usable for models. It runs 32B models at Q4 with room for 8K+ context windows.
What size LLM can I run on an RTX 4090?
Up to 32B parameter models at Q4 quantization. This includes DeepSeek-R1 32B, Qwen 2.5 32B, and larger reasoning models. For 70B models, you need Q2 or dual GPUs, or step up to the RTX 5090.
Is the RTX 4090 worth $2,574 for AI?
For AI-only use, the RTX 3090 at ~$900 used offers 83% of the speed with the same 24GB VRAM. The 4090 is worth it if you also game at 4K or need the absolute fastest 24GB card. For pure AI value, the 3090 wins.
RTX 4090 vs RTX 5090 for AI: which is better?
The RTX 5090 is 39% faster on 8B models (an estimated 145 vs 104 tok/s) with 8GB more VRAM (32 vs 24GB), enabling 70B models. Priced similarly (~$2,500). If buying new in 2026, the 5090 is the clear choice.
Can the RTX 4090 run 70B models?
Not at full quality. A 70B Q4 model needs ~42GB VRAM. The 4090 has 24GB, so you would need Q2 quantization (lower quality) or run with partial CPU offloading (much slower). For 70B, the RTX 5090 (32GB) is recommended.
How fast is a 27B-class model like Qwen3.5 27B Instruct on the RTX 4090?
ModelFit estimates a 32B model on the RTX 4090 runs at roughly 32 tok/s at Q4_K_M. The current 27B-class pick in the catalog is Qwen3.5 27B Instruct (ollama run qwen3.5:27b).
The weekly local-AI refresh
New open-weight models, real Apple Silicon benchmarks, and the one model worth running on your Mac this week. Free, one email a week, unsubscribe anytime.
By subscribing you agree to our Privacy Policy and to receive the weekly email. Unsubscribe anytime.
Want Personalized Recommendations?
Use our interactive wizard to compare models across Apple Silicon and NVIDIA GPUs.