DeskGPU
English

NVIDIA GeForce RTX 5090: which LLMs can it run?

This page was written and translated automatically by AI and checked by software against its sources, with no human editing. Specifications come from the sources listed; memory requirements and speeds are estimates calculated with published formulas. How we use AI

NVIDIA GeForce RTX 5090 has 32 GB of memory (manufacturer figure). Below, every model in the database is checked against it with llama.cpp at a context length of 8,192 tokens (or the model’s maximum if shorter).

Values updated Oct 19, 2026

Specifications

Memory (manufacturer figure)
32 GB
Memory type
GDDR7
Memory bus width
512 bit
Memory bandwidth
1,792 GB/s
Board power (TDP)
575 W
FP16 tensor, dense
Not shown: open sources do not give two independent, agreeing values with a manufacturer reference.
BF16 tensor, dense
Not shown: open sources do not give two independent, agreeing values with a manufacturer reference.
FP8 tensor, dense
Not shown: open sources do not give two independent, agreeing values with a manufacturer reference.
FP4 tensor, dense
Not shown: open sources do not give two independent, agreeing values with a manufacturer reference.
INT8 tensor, dense
Not shown: open sources do not give two independent, agreeing values with a manufacturer reference.
Interface
Not shown: open sources do not give two independent, agreeing values with a manufacturer reference.

Each value lists every source with its own figure; units have been normalized. A manufacturer’s “GB” of GPU memory is counted as GiB (2³⁰ bytes), which is also how AMD’s ROCm table states it. How specifications are checked

Run profile

Run profile: llama.cpp, CUDA back end (NVIDIA discrete GPU), all layers on the GPU, Flash Attention on, micro-batch (-ub) 512, F16 KV cache, context (-c) as shown; “Fits” is judged with llama-server’s default 4 parallel sequences (-np auto, unified KV cache) and “Doesn’t fit” with 1 sequence; idle GPU with no other programs using its memory.

Memory counted as available for “Fits” (C_low): 29.25 GiB per card = 32 GiB nominal − 659 MiB (nominal minus the total that llama.cpp reported for this GPU in public logs) − 2,112 MiB (the largest start-up and runtime use seen in this profile’s public logs), rounded down to a multiple of 64 MiB.

These results apply to the configuration seen in public logs, where llama.cpp reported 32,109 MiB of total memory for this GPU. Settings that reduce usable memory (for example ECC, where the card supports it) are not covered.

Memory used for “Doesn’t fit” (C_high): the full 32 GiB per card. Run profiles and their measurements

Which models fit

“Fits” means llama.cpp's default --fit load check would place all layers on the GPU: our estimate of llama.cpp's buffers, plus its error margin, plus llama.cpp's 1 GiB per-card reserve, is within the memory counted as available. It does not promise that peak memory while generating fits. How results are decided

Context 8,192 tokens (or the model’s maximum if shorter). The last columns show the longest context at that quantization and the bandwidth-based speed ceiling.
Model Parameters Highest quantization that fits Available memory needed for “Fits” Longest context Speed ceiling Result
Qwen3-Coder-Next-Base 79.7B (3.6B active) IQ2_M 27.2 GiB 85,248 ≤ 1,076 tok/s Fits
KAT-Dev-72B-Exp 72.7B IQ2_XXS 28.5 GiB 10,240 ≤ 64.6 tok/s Fits
Llama-3.1-70B-Instruct 70.6B IQ2_M 27.1 GiB 14,592 ≤ 68.0 tok/s Fits
AI21-Jamba2-Mini 51.6B (11.8B active) IQ4_XS 28.3 GiB 59,904 ≤ 274 tok/s Fits
Kimi-Linear-48B-A3B-Base 49.1B (3.1B active) Q4_0 28.7 GiB 64,256 ≤ 909 tok/s Fits
Kimi-Linear-48B-A3B-Instruct 49.1B (3.1B active) IQ4_XS 27.3 GiB 201,984 ≤ 862 tok/s Fits
Hermes-4.3-36B 36.2B Q5_K_M 28.0 GiB 13,056 ≤ 65.9 tok/s Fits
Kimi-Linear-REAP-35B-A3B-Instruct 35.1B (3.1B active) Q5_K_M 25.8 GiB ≥ 262,144 (longer contexts not checked against measured logs) ≤ 766 tok/s Fits
Holo-3.1-35B-A3B 34.7B (2.9B active) Q5_K_M 25.8 GiB 163,072 ≤ 749 tok/s Fits
Holo3-35B-A3B 34.7B (2.9B active) Q5_K_M 25.8 GiB 163,072 ≤ 749 tok/s Fits
Holo4-35B-A3B 34.7B (2.9B active) Q5_K_M 27.9 GiB 71,424 ≤ 585 tok/s Fits
Ising-Calibration-1-35B-A3B 34.7B (2.9B active) Q5_K_M 25.8 GiB 163,072 ≤ 749 tok/s Fits
KAT-Coder-V2.5-Dev 34.7B (2.9B active) Q5_K_M 26.1 GiB 150,272 ≤ 669 tok/s Fits
Qwen3.5-35B-A3B 34.7B (2.9B active) Q5_K_M 27.1 GiB 103,936 ≤ 591 tok/s Fits
Qwen3.5-35B-A3B-Base 34.7B (2.9B active) Q5_K_M 25.9 GiB 161,536 ≤ 738 tok/s Fits
Qwen3.6-35B-A3B 34.7B (2.9B active) Q5_K_M 26.1 GiB 150,272 ≤ 669 tok/s Fits
MiniCPM-V-4.7-35B-A3B 34.7B (2.9B active) Q5_K_M 25.8 GiB 163,072 ≤ 749 tok/s Fits
Laguna-XS-2.1 33.4B (2.8B active) Q5_K_M 25.3 GiB 101,376 ≤ 667 tok/s Fits
DeepSeek-R1-Distill-Qwen-32B 32.8B Q6_K 29.2 GiB 8,448 ≤ 63.1 tok/s Fits
Qwen2.5-32B-Instruct 32.8B Q6_K 29.2 GiB 8,448 ≤ 63.1 tok/s Fits
Qwen2.5-Coder-32B-Instruct 32.8B Q6_K 29.2 GiB 8,448 ≤ 63.1 tok/s Fits
QwQ-32B 32.8B Q6_K 29.2 GiB 8,448 ≤ 63.1 tok/s Fits
AesCode-32B 32.8B Q5_K_M 26.2 GiB 19,712 ≤ 70.5 tok/s Fits
Cosmos-Reason2-32B 32.8B Q6_K 29.2 GiB 8,448 ≤ 63.1 tok/s Fits
FrogBoss-32B-2510 32.8B Q6_K 29.2 GiB 8,448 ≤ 63.1 tok/s Fits
Nemotron-Terminal-32B 32.8B Q6_K 29.2 GiB 8,448 ≤ 63.1 tok/s Fits
Qwen3-32B 32.8B Q6_K 29.2 GiB 8,448 ≤ 63.1 tok/s Fits
Qwen3-VL-32B-Instruct 32.8B Q6_K 29.2 GiB 8,448 ≤ 63.1 tok/s Fits
Qwen3-VL-32B-Thinking 32.8B Q6_K 29.2 GiB 8,448 ≤ 63.1 tok/s Fits
SERA-32B 32.8B Q6_K 29.2 GiB 8,448 ≤ 63.1 tok/s Fits
WebWorld-32B 32.8B Q6_K 29.2 GiB 8,448 ≤ 63.1 tok/s Fits
EXAONE-4.5-33B 32.5B Q6_K 29.0 GiB 9,216 ≤ 65.4 tok/s Fits
Olmo-3.1-32B-Instruct 32.2B Q6_K 28.9 GiB 9,472 ≤ 65.5 tok/s Fits
Olmo-3.1-32B-Instruct-DPO 32.2B Q6_K 28.9 GiB 9,472 ≤ 65.5 tok/s Fits
Olmo-3.1-32B-Think 32.2B Q6_K 28.9 GiB 9,472 ≤ 65.5 tok/s Fits
Olmo-3-1125-32B 32.2B Q6_K 28.9 GiB 9,472 ≤ 65.5 tok/s Fits
Olmo-3-32B-Think 32.2B Q6_K 28.9 GiB 9,472 ≤ 65.5 tok/s Fits
Olmo-3-32B-Think-DPO 32.2B Q6_K 28.9 GiB 9,472 ≤ 65.5 tok/s Fits
Olmo-3-32B-Think-SFT 32.2B Q6_K 28.9 GiB 9,472 ≤ 65.5 tok/s Fits
Holotron4-30B-A3B 31.6B (3.2B active) Q5_K_M 26.8 GiB ≥ 262,144 (longer contexts not checked against measured logs) ≤ 657 tok/s Fits
Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16 31.6B (3.2B active) Q4_K_M 23.3 GiB 262,144 ≤ 603 tok/s Fits
Nemotron-Cascade-2-30B-A3B 31.6B (3.2B active) Q5_K_M 27.0 GiB 262,144 ≤ 618 tok/s Fits
NVIDIA-Nemotron-3.5-Lightning-30B-A3B-Base-BF16 31.6B (3.2B active) Q5_K_M 26.6 GiB 262,144 ≤ 660 tok/s Fits
NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 31.6B (3.2B active) Q5_K_M 27.0 GiB 262,144 ≤ 618 tok/s Fits
NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 31.6B (3.2B active) Q5_K_M 26.9 GiB 262,144 ≤ 632 tok/s Fits
NVIDIA-Nemotron-Labs-3-Elastic-30B-A3B-BF16 31.6B (3.2B active) Q5_K_M 26.8 GiB 262,144 ≤ 657 tok/s Fits
gemma-4-31B 30.7B Q5_K_M 27.3 GiB 31,744 ≤ 76.8 tok/s Fits
gemma-4-31B-it 30.7B Q5_K_M 27.1 GiB 33,792 ≤ 77.4 tok/s Fits
Ising-Calibration-1.5-31B-BF16 30.7B Q5_K_M 28.1 GiB 22,528 ≤ 74.3 tok/s Fits
Holo2-30B-A3B 30.5B (3.0B active) Q6_K 26.5 GiB 36,608 ≤ 536 tok/s Fits
Qwen3-30B-A3B 30.5B (3.0B active) Q6_K 26.5 GiB 36,608 ≤ 536 tok/s Fits
Qwen3-30B-A3B-Instruct-2507 30.5B (3.0B active) Q6_K 26.5 GiB 36,608 ≤ 536 tok/s Fits
Qwen3-Coder-30B-A3B-Instruct 30.5B (3.0B active) Q6_K 26.5 GiB 36,608 ≤ 536 tok/s Fits
Qwen3-VL-30B-A3B-Instruct 30.5B (3.0B active) Q6_K 26.5 GiB 36,608 ≤ 536 tok/s Fits
UI-Venus-1.5-30B-A3B 30.5B (3.0B active) Q6_K 26.5 GiB 36,608 ≤ 536 tok/s Fits
North-Mini-Code-1.0 30.5B (3.3B active) Q6_K 28.2 GiB 20,992 ≤ 509 tok/s Fits
GLM-4.7-Flash 29.9B (3.6B active) Q6_K 25.7 GiB 72,448 ≤ 512 tok/s Fits
granite-4.2-30b 29.3B Q6_K 26.6 GiB 18,176 ≤ 69.4 tok/s Fits
granite-4.1-30b 28.9B Q6_K 26.6 GiB 18,176 ≤ 69.4 tok/s Fits
Muse-Glimmer-30B 27.9B Q6_K 23.7 GiB 131,072 ≤ 79.7 tok/s Fits
gemma-2-27b-it 27.2B Q6_K 26.5 GiB 8,192 ≤ 72.7 tok/s Fits
gemma-3-27b-it 27.0B Q6_K 25.9 GiB 49,152 ≤ 77.0 tok/s Fits
translategemma-27b-it 27.0B Q6_K 25.9 GiB 49,152 ≤ 77.0 tok/s Fits
Fara1.5-27B 26.9B Q6_K 24.4 GiB 81,408 ≤ 78.9 tok/s Fits
Holo4-27B 26.9B Q6_K 23.3 GiB 98,304 ≤ 83.1 tok/s Fits
Qwen3.5-27B 26.9B Q6_K 23.6 GiB 92,672 ≤ 81.7 tok/s Fits
Qwen3.6-27B 26.9B Q6_K 23.7 GiB 91,648 ≤ 81.4 tok/s Fits
Qwen3.8-27B 26.9B Q6_K 24.8 GiB 75,520 ≤ 77.6 tok/s Fits
Qwen3.8-3.6-27B-blend 26.9B Q6_K 24.8 GiB 75,520 ≤ 77.6 tok/s Fits
UI-Mate-27B 26.9B Q6_K 24.4 GiB 81,408 ≤ 78.9 tok/s Fits
UI-Mate-democua-27B 26.9B Q6_K 23.3 GiB 98,304 ≤ 83.1 tok/s Fits
Trinity-Mini 26.1B (3.1B active) Q8_0 28.8 GiB 33,280 ≤ 510 tok/s Fits
Trinity-Mini-Base 26.1B (3.1B active) Q6_K 22.7 GiB 131,072 ≤ 646 tok/s Fits
Trinity-Mini-Base-Pre-Anneal 26.1B (3.1B active) Q6_K 22.4 GiB 4,096 ≤ 662 tok/s Fits
gemma-4-26B-A4B 25.2B (3.8B active) Q8_0 29.0 GiB 21,248 ≤ 401 tok/s Fits
gemma-4-26B-A4B-it 25.2B (3.8B active) Q8_0 29.0 GiB 21,248 ≤ 401 tok/s Fits
Qwen3-Coder-REAP-25B-A3B 24.9B (3.0B active) Q8_0 27.7 GiB 23,552 ≤ 441 tok/s Fits
LFM2.5-24B-A2B 24.0B (2.5B active) Q8_0 26.4 GiB 131,072 ≤ 642 tok/s Fits
LFM2-24B-A2B 23.8B (2.3B active) Q8_0 26.3 GiB 128,000 ≤ 675 tok/s Fits
Devstral-Small-2-24B-Instruct-2512 23.6B Q8_0 26.5 GiB 25,344 ≤ 69.8 tok/s Fits
Mistral-Small-3.2-24B-Instruct-2506 23.6B Q8_0 26.5 GiB 25,344 ≤ 69.8 tok/s Fits
GLM-4.7-Flash-REAP-23B-A3B 23.0B (3.6B active) Q8_0 25.4 GiB 78,080 ≤ 421 tok/s Fits
EuroLLM-22B-2512 22.6B Q6_K 20.6 GiB 32,768 ≤ 90.8 tok/s Fits
EuroLLM-22B-Instruct-2512 22.6B Q8_0 25.8 GiB 23,808 ≤ 71.6 tok/s Fits
gpt-oss-20b 20.9B (3.6B active) Q8_0 12.9 GiB 131,072 ≤ 645 tok/s Fits
OptiMind-SFT 20.9B (3.6B active) Q6_K 22.8 GiB 131,072 ≤ 450 tok/s Fits
DeepSeek-R1-Distill-Qwen-14B 14.8B Q8_0 17.5 GiB 68,608 ≤ 109 tok/s Fits
Qwen2.5-14B-Instruct 14.8B Q8_0 17.5 GiB 32,768 ≤ 109 tok/s Fits
ContextPilot-14B 14.8B Q8_0 17.2 GiB 40,960 ≤ 111 tok/s Fits
FrogMini-14B-2510 14.8B Q6_K 13.9 GiB 40,960 ≤ 140 tok/s Fits
JanusCoder-14B 14.8B Q8_0 17.2 GiB 32,768 ≤ 111 tok/s Fits
Nemotron-Cascade-14B-Thinking 14.8B Q8_0 17.2 GiB 32,768 ≤ 111 tok/s Fits
Nemotron-Terminal-14B 14.8B Q6_K 13.9 GiB 40,960 ≤ 140 tok/s Fits
NousCoder-14B 14.8B Q8_0 17.2 GiB 81,920 ≤ 111 tok/s Fits
Qwen3-14B 14.8B Q8_0 17.2 GiB 40,960 ≤ 111 tok/s Fits
Qwen3-Nemotron-14B-BRRM 14.8B Q6_K 13.9 GiB 40,960 ≤ 140 tok/s Fits
WebWorld-14B 14.8B Q6_K 13.9 GiB 40,960 ≤ 140 tok/s Fits
phi-4 14.7B Q8_0 17.7 GiB 16,384 ≤ 107 tok/s Fits
Apriel-1.6-15b-Thinker 14.4B Q8_0 17.2 GiB 69,888 ≤ 110 tok/s Fits
Ministral-3-14B-Instruct-2512 13.5B BF16 27.8 GiB 17,152 ≤ 66.3 tok/s Fits
Ministral-3-14B-Reasoning-2512 13.5B BF16 27.8 GiB 17,152 ≤ 66.3 tok/s Fits
Mistral-Nemo-Instruct-2407 12.2B F16 25.3 GiB 32,512 ≤ 73.2 tok/s Fits
Mellum2.1-12B-A2.5B-Thinking 12.1B (2.2B active) BF16 25.0 GiB 131,072 ≤ 390 tok/s Fits
Mellum2-12B-A2.5B-Base 12.1B (2.2B active) Q6_K 12.1 GiB 131,072 ≤ 853 tok/s Fits
Mellum2-12B-A2.5B-Instruct 12.1B (2.2B active) BF16 25.0 GiB 131,072 ≤ 390 tok/s Fits
Mellum2-12B-A2.5B-Instruct-SFT 12.1B (2.2B active) Q6_K 12.1 GiB 131,072 ≤ 853 tok/s Fits
Mellum2-12B-A2.5B-Thinking 12.1B (2.2B active) BF16 25.0 GiB 131,072 ≤ 390 tok/s Fits
Mellum2-12B-A2.5B-Thinking-SFT 12.1B (2.2B active) Q6_K 12.1 GiB 131,072 ≤ 853 tok/s Fits
gemma-4-12B 11.9B BF16 26.5 GiB 159,488 ≤ 73.8 tok/s Fits
gemma-4-12B-it 11.9B BF16 26.5 GiB 159,488 ≤ 73.8 tok/s Fits
gemma-3-12b-it 11.8B BF16 26.6 GiB 47,616 ≤ 73.4 tok/s Fits
translategemma-12b-it 11.8B Q8_0 15.8 GiB 131,072 ≤ 134 tok/s Fits
Bielik-11B-v3.0-Instruct 11.2B BF16 24.6 GiB 31,232 ≤ 75.4 tok/s Fits
AutoGLM-Phone-9B 9.4B F16 18.9 GiB 65,536 ≤ 100 tok/s Fits
AutoGLM-Phone-9B-Multilingual 9.4B Q6_K 9.2 GiB 65,536 ≤ 222 tok/s Fits
GLM-4.6V-Flash 9.4B BF16 18.9 GiB 131,072 ≤ 100 tok/s Fits
gemma-2-9b-it 9.2B Q8_0 14.8 GiB 8,192 ≤ 140 tok/s Fits
EuroLLM-9B-Instruct-2512 9.2B F16 19.6 GiB 32,768 ≤ 96.0 tok/s Fits
Fara1.5-9B 9.0B BF16 17.5 GiB 262,144 ≤ 111 tok/s Fits
Holo-3.1-9B 9.0B F16 17.5 GiB 262,144 ≤ 111 tok/s Fits
MiMo-V2.6-Distill-Qwen-9B 9.0B BF16 17.5 GiB 262,144 ≤ 111 tok/s Fits
Qwen3.5-9B 9.0B BF16 17.5 GiB 262,144 ≤ 111 tok/s Fits
Qwen3.5-9B-Base 9.0B F16 17.5 GiB 262,144 ≤ 111 tok/s Fits
UI-Mate-9B 9.0B BF16 17.5 GiB 262,144 ≤ 111 tok/s Fits
UI-Venus-2-9B 9.0B BF16 17.5 GiB 262,144 ≤ 111 tok/s Fits
NVIDIA-Nemotron-Nano-9B-v2 8.9B BF16 18.3 GiB 131,072 ≤ 107 tok/s Fits
granite-4.1-8b 8.8B BF16 19.0 GiB 70,912 ≤ 99.0 tok/s Fits
granite-4.2-8b 8.8B BF16 19.0 GiB 70,912 ≤ 99.0 tok/s Fits
Ministral-3-8B-Instruct-2512 8.5B BF16 18.0 GiB 89,344 ≤ 105 tok/s Fits
Ministral-3-8B-Reasoning-2512 8.5B BF16 18.0 GiB 89,344 ≤ 105 tok/s Fits
LFM2.5-8B-A1B 8.5B (1.7B active) BF16 18.0 GiB 128,000 ≤ 516 tok/s Fits
granite-guardian-4.1-8b 8.4B BF16 19.0 GiB 70,912 ≤ 99.0 tok/s Fits
LFM2-8B-A1B 8.3B (1.6B active) F16 17.7 GiB 128,000 ≤ 557 tok/s Fits
rnj-1-instruct 8.3B BF16 18.6 GiB 32,768 ≤ 101 tok/s Fits
AesCode-8B 8.2B BF16 17.3 GiB 89,344 ≤ 110 tok/s Fits
ArmorOCR 8.2B Q8_0 10.4 GiB 136,704 ≤ 194 tok/s Fits
AstaBrief_8B 8.2B Q6_K 8.6 GiB 40,960 ≤ 242 tok/s Fits
ContextPilot-8B 8.2B Q6_K 8.6 GiB 40,960 ≤ 242 tok/s Fits
Cosmos-Reason2-8B 8.2B Q6_K 8.6 GiB 148,992 ≤ 242 tok/s Fits
DeepSeek-R1-0528-Qwen3-8B 8.2B BF16 17.3 GiB 89,344 ≤ 110 tok/s Fits
Holo2-8B 8.2B Q6_K 8.6 GiB 148,992 ≤ 242 tok/s Fits
intent-aware-lfqa-qwen3-8b-baseline 8.2B F16 17.3 GiB 40,960 ≤ 110 tok/s Fits
intent-aware-lfqa-qwen3-8b-intent-explicit 8.2B F16 17.3 GiB 40,960 ≤ 110 tok/s Fits
intent-aware-lfqa-qwen3-8b-intent-implicit 8.2B F16 17.3 GiB 40,960 ≤ 110 tok/s Fits
intent-aware-lfqa-qwen3-8b-multiview 8.2B F16 17.3 GiB 40,960 ≤ 110 tok/s Fits
JanusCoder-8B 8.2B BF16 17.3 GiB 32,768 ≤ 110 tok/s Fits
MathForm-8B 8.2B F16 17.3 GiB 40,960 ≤ 110 tok/s Fits
Nemotron-Cascade-8B 8.2B BF16 17.3 GiB 32,768 ≤ 110 tok/s Fits
Nemotron-Cascade-8B-Thinking 8.2B BF16 17.3 GiB 32,768 ≤ 110 tok/s Fits
Nemotron-Orchestrator-8B 8.2B BF16 17.3 GiB 40,960 ≤ 110 tok/s Fits
Nemotron-Terminal-8B 8.2B Q6_K 8.6 GiB 40,960 ≤ 242 tok/s Fits
Qwen3-8B 8.2B Q8_0 10.4 GiB 40,960 ≤ 194 tok/s Fits
Qwen3-Nemotron-8B-BRRM 8.2B Q6_K 8.6 GiB 40,960 ≤ 242 tok/s Fits
Qwen3-VL-8B-Instruct 8.2B BF16 17.3 GiB 89,344 ≤ 110 tok/s Fits
Qwen3-VL-8B-Thinking 8.2B BF16 17.3 GiB 89,344 ≤ 110 tok/s Fits
RLVR-8B-0926 8.2B Q6_K 8.6 GiB 148,992 ≤ 242 tok/s Fits
Step3-VL-10B-Base 8.2B Q6_K 8.6 GiB 40,960 ≤ 242 tok/s Fits
UI-Venus-1.5-8B 8.2B Q6_K 8.6 GiB 148,992 ≤ 242 tok/s Fits
WebWorld-8B 8.2B F16 17.3 GiB 40,960 ≤ 110 tok/s Fits
ZwZ-8B 8.2B Q6_K 8.6 GiB 148,992 ≤ 242 tok/s Fits
MiniCPM-o-4_5 8.2B F16 17.3 GiB 40,960 ≤ 110 tok/s Fits
granite-3.3-8b-instruct 8.2B F16 18.6 GiB 73,472 ≤ 101 tok/s Fits
granite-guardian-3.2-8b-factuality-detection 8.2B F16 18.6 GiB 73,472 ≤ 101 tok/s Fits
intent-aware-lfqa-llama3-8b-baseline 8.0B F16 17.0 GiB 101,632 ≤ 111 tok/s Fits
intent-aware-lfqa-llama3-8b-intent-explicit 8.0B F16 17.0 GiB 101,632 ≤ 111 tok/s Fits
intent-aware-lfqa-llama3-8b-intent-implicit 8.0B F16 17.0 GiB 101,632 ≤ 111 tok/s Fits
intent-aware-lfqa-llama3-8b-multiview 8.0B F16 17.0 GiB 101,632 ≤ 111 tok/s Fits
DeepSeek-R1-Distill-Llama-8B 8.0B BF16 17.0 GiB 101,632 ≤ 111 tok/s Fits
HARC-Llama-3.1-8B-Instruct 8.0B Q6_K 8.4 GiB 131,072 ≤ 248 tok/s Fits
Llama-3.1-8B-Instruct 8.0B BF16 17.0 GiB 101,632 ≤ 111 tok/s Fits
AprielGuard 7.9B Q6_K 7.9 GiB 131,072 ≤ 266 tok/s Fits
MiMo-Embodied-7B 7.6B Q6_K 8.1 GiB 128,000 ≤ 258 tok/s Fits
DeepSeek-R1-Distill-Qwen-7B 7.6B F16 15.6 GiB 131,072 ≤ 123 tok/s Fits
Fara-7B 7.6B BF16 15.6 GiB 128,000 ≤ 123 tok/s Fits
HARC-Qwen2.5-7B-Instruct 7.6B Q6_K 7.5 GiB 32,768 ≤ 286 tok/s Fits
JanusCoderV-7B 7.6B BF16 15.6 GiB 128,000 ≤ 123 tok/s Fits
olmOCR-2-7B-1025 7.6B BF16 15.6 GiB 128,000 ≤ 123 tok/s Fits
Qwen2.5-7B-Instruct 7.6B Q8_0 9.1 GiB 32,768 ≤ 224 tok/s Fits
Qwen2.5-Coder-7B-Instruct 7.6B Q8_0 9.1 GiB 32,768 ≤ 224 tok/s Fits
Qwen2.5-VL-7B-Instruct 7.6B BF16 15.6 GiB 128,000 ≤ 123 tok/s Fits
Qwen2.5-VL-7B-Surg-CholecT50 7.6B F16 15.6 GiB 128,000 ≤ 123 tok/s Fits
Spatial-SSRL-7B 7.6B Q6_K 7.5 GiB 128,000 ≤ 286 tok/s Fits
X-Reasoner-7B 7.6B Q6_K 7.5 GiB 128,000 ≤ 286 tok/s Fits
ZwZ-7B 7.6B Q6_K 7.5 GiB 128,000 ≤ 286 tok/s Fits
gemma-4-E4B 7.5B F16 11.0 GiB 131,072 ≤ 188 tok/s Fits
gemma-4-E4B-it 7.5B BF16 11.0 GiB 131,072 ≤ 188 tok/s Fits
Bielik-Minitron-7B-v3.0-Instruct 7.5B Q8_0 10.2 GiB 32,768 ≤ 196 tok/s Fits
ContextPilot-E4B 7.5B F16 10.9 GiB 131,072 ≤ 190 tok/s Fits
Olmo-3.1-7B-RL-Zero-Code 7.3B Q6_K 11.0 GiB 65,536 ≤ 215 tok/s Fits
Olmo-3.1-7B-RL-Zero-Math 7.3B F16 19.0 GiB 60,672 ≤ 109 tok/s Fits
Olmo-3-7B-Instruct 7.3B BF16 19.0 GiB 60,672 ≤ 109 tok/s Fits
Olmo-3-7B-Instruct-DPO 7.3B Q6_K 11.0 GiB 65,536 ≤ 215 tok/s Fits
Olmo-3-7B-Instruct-SFT 7.3B Q6_K 11.0 GiB 65,536 ≤ 215 tok/s Fits
Olmo-3-7B-RL-Zero-Code 7.3B Q6_K 11.0 GiB 65,536 ≤ 215 tok/s Fits
Olmo-3-7B-RL-Zero-General 7.3B Q6_K 11.0 GiB 65,536 ≤ 215 tok/s Fits
Olmo-3-7B-RL-Zero-IF 7.3B Q6_K 11.0 GiB 65,536 ≤ 215 tok/s Fits
Olmo-3-7B-RL-Zero-Math 7.3B Q6_K 11.0 GiB 65,536 ≤ 215 tok/s Fits
Olmo-3-7B-RL-Zero-Mix 7.3B Q6_K 11.0 GiB 65,536 ≤ 215 tok/s Fits
Olmo-3-7B-Think 7.3B BF16 19.0 GiB 60,672 ≤ 109 tok/s Fits
Olmo-3-7B-Think-DPO 7.3B Q6_K 11.0 GiB 65,536 ≤ 215 tok/s Fits
Olmo-3-7B-Think-SFT 7.3B Q6_K 11.0 GiB 65,536 ≤ 215 tok/s Fits
Mistral-7B-Instruct-v0.2 7.2B Q6_K 8.1 GiB 32,768 ≤ 259 tok/s Fits
Trinity-Nano-Base 6.1B (819M active) Q6_K 6.7 GiB 131,072 ≤ 1,990 tok/s Fits
Trinity-Nano-Base-Pre-Anneal 6.1B (819M active) Q6_K 6.4 GiB 4,096 ≤ 2,129 tok/s Fits
Trinity-Nano-Preview 6.1B (819M active) BF16 13.5 GiB 131,072 ≤ 964 tok/s Fits
Youtu-VL-4B-Instruct 4.9B Q6_K 5.9 GiB 32,768 ≤ 406 tok/s Fits
Holo-3.1-4B 4.8B F16 10.2 GiB 262,144 ≤ 206 tok/s Fits
gemma-4-E2B 4.6B Q6_K 3.5 GiB 131,072 ≤ 915 tok/s Fits
gemma-4-E2B-it 4.6B BF16 6.1 GiB 131,072 ≤ 385 tok/s Fits
AgentCPM-Explore 4.4B Q6_K 5.7 GiB 168,448 ≤ 397 tok/s Fits
CapRL-Qwen3VL-4B 4.4B F16 10.4 GiB 136,704 ≤ 194 tok/s Fits
Holo2-4B 4.4B F16 10.4 GiB 136,704 ≤ 194 tok/s Fits
intent-aware-lfqa-qwen3-4b-baseline 4.4B F16 10.4 GiB 40,960 ≤ 194 tok/s Fits
intent-aware-lfqa-qwen3-4b-intent-explicit 4.4B F16 10.4 GiB 40,960 ≤ 194 tok/s Fits
intent-aware-lfqa-qwen3-4b-intent-implicit 4.4B F16 10.4 GiB 40,960 ≤ 194 tok/s Fits
intent-aware-lfqa-qwen3-4b-multiview 4.4B F16 10.4 GiB 40,960 ≤ 194 tok/s Fits
Spatial-SSRL-Qwen3VL-4B 4.4B F16 10.4 GiB 136,704 ≤ 194 tok/s Fits
ZwZ-4B 4.4B Q6_K 5.7 GiB 168,448 ≤ 397 tok/s Fits
Fara1.5-4B 4.2B BF16 10.2 GiB 262,144 ≤ 206 tok/s Fits
FrogNano-4B-2609 4.2B BF16 10.2 GiB 262,144 ≤ 206 tok/s Fits
Qwen3.5-4B 4.2B BF16 10.2 GiB 262,144 ≤ 206 tok/s Fits
Qwen3.5-4B-Base 4.2B F16 10.2 GiB 262,144 ≤ 206 tok/s Fits
UserRepGen-4B 4.2B BF16 10.2 GiB 262,144 ≤ 206 tok/s Fits
Riva-Translate-4B-Instruct-v1.1 4.2B Q6_K 6.0 GiB 8,192 ≤ 374 tok/s Fits
GELab-Zero-4B-preview 4.0B BF16 10.4 GiB 136,704 ≤ 194 tok/s Fits
Qwen3-4B 4.0B Q8_0 6.7 GiB 40,960 ≤ 327 tok/s Fits
Qwen3-4B-Instruct-2507 4.0B F16 10.4 GiB 136,704 ≤ 194 tok/s Fits
Qwen3-VL-4B-Instruct 4.0B BF16 10.4 GiB 136,704 ≤ 194 tok/s Fits
Qwen3-VL-4B-Thinking 4.0B BF16 10.4 GiB 136,704 ≤ 194 tok/s Fits
NVIDIA-Nemotron-3-Nano-4B-BF16 4.0B BF16 8.7 GiB 262,144 ≤ 247 tok/s Fits
gemma-3-4b-it 3.9B BF16 9.9 GiB 131,072 ≤ 223 tok/s Fits
medgemma-1.5-4b-it 3.9B BF16 9.9 GiB 131,072 ≤ 223 tok/s Fits
Nemotron-3-Content-Safety 3.9B Q6_K 5.4 GiB 131,072 ≤ 516 tok/s Fits
translategemma-4b-it 3.9B F16 9.9 GiB 131,072 ≤ 223 tok/s Fits
Phi-4-mini-instruct 3.8B BF16 10.0 GiB 131,072 ≤ 205 tok/s Fits
Apertus-v1.1-4B 3.8B Q6_K 4.4 GiB 4,096 ≤ 558 tok/s Fits
Phi-3.5-mini-instruct 3.8B Q8_0 8.3 GiB 62,208 ≤ 250 tok/s Fits
granite-4.2-3b 3.7B BF16 8.6 GiB 131,072 ≤ 240 tok/s Fits
BitCPM-CANN-3B-unquantized 3.6B Q6_K 4.3 GiB 32,768 ≤ 583 tok/s Fits
kanana-2-3b-instruct 3.5B F16 8.6 GiB 32,768 ≤ 241 tok/s Fits
Ministral-3-3B-Instruct-2512 3.4B BF16 8.9 GiB 199,680 ≤ 232 tok/s Fits
Ministral-3-3B-Reasoning-2512 3.4B BF16 8.9 GiB 199,680 ≤ 232 tok/s Fits
granite-vision-4.1-4b 3.4B BF16 8.6 GiB 131,072 ≤ 240 tok/s Fits
granite-4.1-3b 3.4B BF16 8.6 GiB 131,072 ≤ 240 tok/s Fits
granite-guardian-4.0-3b-toxicity-ja 3.4B Q6_K 4.7 GiB 131,072 ≤ 517 tok/s Fits
Qwen2.5-3B-Instruct 3.4B Q8_0 4.8 GiB 32,768 ≤ 500 tok/s Fits
tiny-aya-earth 3.3B BF16 8.7 GiB ≥ 262,144 (longer contexts not checked against measured logs) ≤ 253 tok/s Fits
tiny-aya-fire 3.3B BF16 8.7 GiB ≥ 262,144 (longer contexts not checked against measured logs) ≤ 253 tok/s Fits
tiny-aya-global 3.3B BF16 8.7 GiB ≥ 262,144 (longer contexts not checked against measured logs) ≤ 253 tok/s Fits
tiny-aya-water 3.3B BF16 8.7 GiB ≥ 262,144 (longer contexts not checked against measured logs) ≤ 253 tok/s Fits
Llama-3.2-3B-Instruct 3.2B BF16 8.5 GiB 131,072 ≤ 243 tok/s Fits
AI21-Jamba-Reasoning-3B 3.2B BF16 7.3 GiB 262,144 ≤ 295 tok/s Fits
AI21-Jamba2-3B 3.2B BF16 7.3 GiB 262,144 ≤ 295 tok/s Fits
kanana-2-3b-base 3.2B Q6_K 4.9 GiB 32,768 ≤ 487 tok/s Fits
SmolLM3-3B 3.1B BF16 7.9 GiB 65,536 ≤ 265 tok/s Fits
SmolLM3-3B-GSM8K-SFT 3.1B Q6_K 4.3 GiB 65,536 ≤ 573 tok/s Fits
LFM2-2.6B-Exp 2.7B F16 6.5 GiB 128,000 ≤ 340 tok/s Fits
d1-3B 2.7B BF16 6.7 GiB 32,768 ≤ 324 tok/s Fits
LFM2.5-2.6B 2.7B BF16 6.7 GiB 131,072 ≤ 324 tok/s Fits
LFM2.5-2.6B-Base 2.7B Q6_K 3.6 GiB 128,000 ≤ 763 tok/s Fits
LFM2.5-VL-3B 2.7B BF16 6.7 GiB 32,768 ≤ 324 tok/s Fits
EuroMoE-2.6B-A0.6B-Instruct-2512 2.6B (498M active) Q6_K 3.6 GiB 32,768 ≤ 2,915 tok/s Fits
LFM2-2.6B-Longevity 2.6B F16 6.5 GiB 128,000 ≤ 340 tok/s Fits
LFM2-2.6B-Transcript 2.6B BF16 6.5 GiB 128,000 ≤ 340 tok/s Fits
LFM2-VL-3B 2.6B BF16 6.5 GiB 128,000 ≤ 340 tok/s Fits
MiniCPM5-2B 2.5B BF16 6.0 GiB 131,072 ≤ 369 tok/s Fits
MiniCPM5-2B-Base 2.5B Q6_K 3.4 GiB ≥ 262,144 (longer contexts not checked against measured logs) ≤ 815 tok/s Fits
CapRL-Qwen3VL-2B 2.0B F16 5.6 GiB 214,528 ≤ 409 tok/s Fits
HiPO-1.7B 2.0B Q6_K 3.6 GiB 40,960 ≤ 762 tok/s Fits
UI-Venus-1.5-2B 2.0B F16 5.6 GiB 214,528 ≤ 409 tok/s Fits
ZwZ-2B 2.0B F16 5.6 GiB 214,528 ≤ 409 tok/s Fits
Youtu-LLM-2B 2.0B Q6_K 3.2 GiB 131,072 ≤ 933 tok/s Fits
Youtu-LLM-2B-Base 2.0B Q6_K 3.2 GiB 131,072 ≤ 933 tok/s Fits
Qwen3.5-2B 1.9B BF16 5.4 GiB 262,144 ≤ 464 tok/s Fits
Qwen3.5-2B-Base 1.9B Q6_K 3.2 GiB 262,144 ≤ 1,089 tok/s Fits
Qwen3-1.7B 1.7B BF16 5.6 GiB 40,960 ≤ 409 tok/s Fits
Qwen3-VL-2B-Instruct 1.7B BF16 5.6 GiB 214,528 ≤ 409 tok/s Fits
Qwen3-VL-2B-Thinking 1.7B BF16 5.6 GiB 214,528 ≤ 409 tok/s Fits
SingGuard-2b 1.7B F16 5.6 GiB 214,528 ≤ 409 tok/s Fits
granite-4.0-1b 1.6B BF16 5.1 GiB 131,072 ≤ 455 tok/s Fits
Apertus-v1.1-1.5B 1.5B F16 3.8 GiB 4,096 ≤ 685 tok/s Fits
Apertus-v1.1-1.5B-Instruct 1.5B F16 3.8 GiB 4,096 ≤ 685 tok/s Fits
granite-4.0-h-1b 1.5B BF16 4.4 GiB 131,072 ≤ 599 tok/s Fits
Llama-3.2-1B-Instruct 1.2B BF16 4.0 GiB 131,072 ≤ 654 tok/s Fits
LFM2.5-1.2B-Base 1.2B BF16 3.7 GiB 128,000 ≤ 734 tok/s Fits
LFM2.5-1.2B-Instruct 1.2B BF16 3.7 GiB 128,000 ≤ 734 tok/s Fits
LFM2.5-1.2B-JP 1.2B F16 3.7 GiB 128,000 ≤ 734 tok/s Fits
LFM2.5-1.2B-JP-202606 1.2B F16 3.7 GiB 128,000 ≤ 734 tok/s Fits
LFM2.5-1.2B-Thinking 1.2B BF16 3.7 GiB 128,000 ≤ 734 tok/s Fits
LFM2.5-VL-1.6B 1.2B BF16 3.7 GiB 128,000 ≤ 734 tok/s Fits
LFM2.5-VL-1.6B-Extract 1.2B F16 3.7 GiB 128,000 ≤ 734 tok/s Fits
MiniCPM5-1B 1.1B Q6_K 2.2 GiB 131,072 ≤ 1,940 tok/s Fits
MiniCPM5-1B-Base 1.1B Q6_K 2.2 GiB 131,072 ≤ 1,940 tok/s Fits
MiniCPM5-1B-SFT 1.1B Q6_K 2.2 GiB 131,072 ≤ 1,940 tok/s Fits
Holo-3.1-0.8B 1.0B Q6_K 2.3 GiB 262,144 ≤ 2,491 tok/s Fits
MiniCPM-V-4.7-1B 1.0B F16 3.2 GiB 262,144 ≤ 1,116 tok/s Fits
gemma-3-1b-it 1.0B BF16 3.6 GiB 32,768 ≤ 876 tok/s Fits
Qwen3.5-0.8B 752M BF16 3.2 GiB 262,144 ≤ 1,116 tok/s Fits
Qwen3.5-0.8B-Base 752M Q6_K 2.3 GiB 262,144 ≤ 2,491 tok/s Fits
MiniCPM-V-4.6 752M F16 3.2 GiB 262,144 ≤ 1,116 tok/s Fits
MiniCPM-V-4.6-Thinking 752M F16 3.2 GiB 262,144 ≤ 1,116 tok/s Fits
Qwen3-0.6B 752M BF16 3.4 GiB 40,960 ≤ 841 tok/s Fits
HunyuanOCR 539M F16 3.1 GiB 32,768 ≤ 951 tok/s Fits
LFM2.5-350M 354M BF16 2.1 GiB 128,000 ≤ 2,213 tok/s Fits
LFM2.5-Encoder-350M-Diffusion 354M Q6_K 1.7 GiB 128,000 ≤ 4,575 tok/s Fits
LFM2.5-VL-450M 354M BF16 2.1 GiB 128,000 ≤ 2,213 tok/s Fits
LFM2.5-VL-450M-Extract 354M F16 2.1 GiB 128,000 ≤ 2,213 tok/s Fits
granite-4.0-350m 352M BF16 2.2 GiB 32,768 ≤ 1,907 tok/s Fits
granite-4.0-h-350m 340M BF16 2.1 GiB 32,768 ≤ 2,507 tok/s Fits
functiongemma-270m-it 268M BF16 2.1 GiB 32,768 ≤ 3,148 tok/s Fits
LFM2.5-230M 230M BF16 1.8 GiB 128,000 ≤ 3,199 tok/s Fits

Borderline models

Within the estimate error range: it may fit or it may not. Try a shorter context or a lower quantization.

Context 8,192 tokens. No quantization of these models fits; the one shown is the highest that is borderline.
Model Parameters Highest borderline quantization Available memory needed for “Fits” Result
Qwen3.5-122B-A10B 122B (9.0B active) IQ2_XXS 35.0 GiB Tight
Step-3.5-Flash-REAP-121B-A11B 121B (11.4B active) IQ2_XXS 32.9 GiB Tight
Mistral-Small-4-119B-2603 119B (6.1B active) IQ2_XXS 34.3 GiB Tight
Solar-Open-100B 103B (11.2B active) IQ2_M 35.8 GiB Tight
GLM-4.5-Air-REAP-82B-A12B 81.9B (12.8B active) Q2_K 36.5 GiB Tight
Qwen3-Coder-Next 79.7B (3.6B active) Q2_K 30.4 GiB Tight
Qwen3-Next-80B-A3B-Instruct 79.7B (3.6B active) Q2_K 30.3 GiB Tight
Qwen2.5-72B-Instruct 73.0B Q2_K 30.3 GiB Tight
DeepSeek-R1-Distill-Llama-70B 70.6B Q3_K_M 37.1 GiB Tight
Llama-3.3-70B-Instruct 70.6B Q3_K_M 37.1 GiB Tight
Qwen-AgentWorld-35B-A3B 34.7B (2.9B active) Q8_0 37.6 GiB Tight

Models that don’t fit on one NVIDIA GeForce RTX 5090

Context 8,192 tokens. Even the lowest quantization does not fit on one card; the requirement shown is for that quantization.
Model Parameters Smallest quantization Available memory needed for “Fits” Several identical cards Result
MiniMax-M2.5-REAP-139B-A10B 139B (10.4B active) IQ2_XXS 39.2 GiB 2 identical cards (default even split) are expected to fit. Doesn't fit
INTELLECT-3 107B (12.8B active) IQ2_XXS 39.6 GiB 2 identical cards (default even split) are expected to fit. Doesn't fit
Step-3.5-Flash-REAP-149B-A11B 149B (11.4B active) IQ2_XXS 40.1 GiB 2 identical cards (default even split) are expected to fit. Doesn't fit
Llama-4-Scout-17B-16E-Instruct 108B (16.1B active) Q2_K 41.5 GiB 2 identical cards (default even split) are expected to fit. Doesn't fit
GLM-4.5-Air 107B (12.8B active) Q2_K 45.6 GiB 2 identical cards (default even split) are expected to fit. Doesn't fit
GLM-4.6V 107B (12.8B active) Q2_K 45.6 GiB 2 identical cards (default even split) are expected to fit. Doesn't fit
Step-3.5-Flash 197B (11.5B active) IQ2_XXS 48.9 GiB 2 identical cards (default even split) are expected to fit. Doesn't fit
Devstral-2-123B-Instruct-2512 125B Q2_K 49.3 GiB 2 identical cards (default even split) are expected to fit. Doesn't fit
Mistral-Medium-3.5-128B 125B Q2_K 49.3 GiB 2 identical cards (default even split) are expected to fit. Doesn't fit
NVIDIA-Nemotron-3-Super-120B-A12B-Base-BF16 121B (12.2B active) IQ2_XXS 49.8 GiB 2 identical cards (default even split) are expected to fit. Doesn't fit
NVIDIA-Nemotron-3-Super-120B-A12B-BF16 121B (12.2B active) IQ2_XXS 51.0 GiB 2 identical cards (default even split) are expected to fit. Doesn't fit
gpt-oss-120b 117B (5.1B active) Q2_K 62.8 GiB 3 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
MiniMax-M2.7 229B (10.4B active) IQ2_XXS 62.9 GiB 3 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
Baichuan-M3-235B 235B (21.6B active) IQ2_XXS 63.2 GiB 3 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
Holo2-235B-A22B 235B (21.6B active) IQ2_XXS 63.2 GiB 3 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
Qwen3-Nemotron-235B-A22B-GenRM 235B (21.6B active) IQ2_XXS 63.2 GiB 3 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
command-a-plus-05-2026-bf16 218B (25.0B active) IQ2_XXS 63.6 GiB 3 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 121B (12.2B active) Q4_0 71.8 GiB 3 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
GLM-4.7-REAP-268B-A32B 269B (32.8B active) IQ2_XXS 73.0 GiB 3 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
Step-3.7-Flash 197B (11.5B active) IQ3_XXS 76.1 GiB 3 identical cards (default even split) are expected to fit. Doesn't fit
Hy3 295B (20.2B active) IQ2_XXS 82.2 GiB 3 identical cards (default even split) are expected to fit. Doesn't fit
MiMo-V2.5 309B (14.8B active) IQ2_XXS 82.4 GiB 4 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
GLM-4.7-REAP-218B-A32B 218B (32.8B active) Q2_K 82.9 GiB 4 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
MiniMax-M2 229B (10.4B active) Q2_K 84.9 GiB 4 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
MiniMax-M2.1 229B (10.4B active) Q2_K 84.9 GiB 4 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
MiniMax-M2.5 229B (10.4B active) Q2_K 84.9 GiB 4 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
Qwen3-235B-A22B-Instruct-2507 235B (21.6B active) Q2_K 86.7 GiB 4 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
Qwen3-VL-235B-A22B-Instruct 235B (21.6B active) Q2_K 86.7 GiB 4 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
GLM-5.3-Flash-BF16 313B (16.7B active) IQ2_XXS 92.0 GiB 4 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
Laguna-S-2.1 118B (8.1B active) Q4_K_M 95.9 GiB 8 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
Trinity-Large-Base 399B (12.8B active) IQ2_XXS 96.4 GiB 4 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
Trinity-Large-TrueBase 399B (12.8B active) IQ2_XXS 96.4 GiB 4 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
GLM-4.6-REAP-268B-A32B 269B (32.8B active) Q2_K 100.9 GiB 4 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
Trinity-Large-Thinking 399B (12.8B active) IQ2_XXS 107.1 GiB 5 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
MiMo-V2-Flash 309B (14.8B active) Q2_K 111.6 GiB 5 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
Intern-S2-397B 396B (16.3B active) IQ2_XXS 114.3 GiB 5 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
MiMo-V2.6-Flash-MOPD 309B (14.8B active) Q2_K 124.9 GiB 5 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
MiMo-V2.6-Flash-RL 309B (14.8B active) Q2_K 124.9 GiB 5 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
K-EXAONE-236B-A23B 237B (22.7B active) IQ4_XS 126.3 GiB 6 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
GLM-4.5 353B (32.9B active) Q2_K 130.7 GiB 5 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
GLM-4.7 353B (32.9B active) Q2_K 130.7 GiB 5 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
Qwen3-Coder-REAP-363B-A35B 363B (34.5B active) Q2_K 132.7 GiB 5 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
Qwen3-235B-A22B 235B (21.6B active) Q4_K_M 141.9 GiB 6 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
Qwen3-Coder-480B-A35B-Instruct 480B (34.5B active) Q2_K 174.3 GiB 7 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
Qwen3.5-397B-A17B 396B (16.3B active) Q3_K_M 175.6 GiB 7 identical cards (default even split) are expected to fit. Fewer cards are borderline. Doesn't fit
GLM-5.1 744B (40.8B active) IQ2_XXS 197.6 GiB Borderline with 8 cards. Doesn't fit
DeepSeek-V3.2 671B (36.6B active) Q2_K 242.1 GiB Borderline with 8 cards. Doesn't fit
cogito-671b-v2.1 671B (36.6B active) Q2_K 242.4 GiB Borderline with 8 cards. Doesn't fit
DeepSeek-V3.1 671B (36.6B active) Q2_K 242.4 GiB Borderline with 8 cards. Doesn't fit
Mistral-Large-3-675B-Instruct-2512 673B (39.0B active) Q2_K 243.8 GiB Borderline with 8 cards. Doesn't fit
DeepSeek-V3 671B (36.6B active) Q2_K 257.8 GiB Borderline with 8 cards. Doesn't fit
Kimi-K2.6 1.03T (31.7B active) IQ2_XXS 267.6 GiB Borderline with 8 cards. Doesn't fit
GLM-5 744B (40.8B active) Q2_K 268.9 GiB Borderline with 8 cards. Doesn't fit
MiMo-V2.5-Pro 1.02T (41.0B active) IQ2_XXS 270.4 GiB Borderline with 8 cards. Doesn't fit
DeepSeek-R1 671B (36.6B active) Q2_K 280.3 GiB Borderline with 8 cards. Doesn't fit
Muse-Spark-Open-816B-A42B 820B (44.4B active) Q2_K 301.0 GiB Does not fit even on 8 cards. Doesn't fit
GLM-5.3-Flash 313B (16.7B active) Q8_0 327.9 GiB Does not fit even on 8 cards. Doesn't fit
Ling-1T 1.00T (49.8B active) Q2_K 360.6 GiB Does not fit even on 8 cards. Doesn't fit
Kimi-K2-Instruct 1.03T (31.7B active) Q2_K 367.4 GiB Does not fit even on 8 cards. Doesn't fit
Kimi-K2.5 1.03T (31.7B active) Q2_K 368.0 GiB Does not fit even on 8 cards. Doesn't fit
Kimi-K2-Thinking 1.03T (31.7B active) Q2_K 368.3 GiB Does not fit even on 8 cards. Doesn't fit
MiMo-V2.6-Pro-MOPD 1.02T (41.0B active) Q2_K 411.6 GiB Does not fit even on 8 cards. Doesn't fit
MiMo-V2.6-Pro-RL 1.02T (41.0B active) Q2_K 411.6 GiB Does not fit even on 8 cards. Doesn't fit
Beam-1.0-501B-A23B 501B (22.7B active) Q8_0 523.6 GiB Does not fit even on 8 cards. Doesn't fit
GLM-5.2 744B (40.8B active) Q8_0 776.4 GiB Does not fit even on 8 cards. Doesn't fit
GLM-5.3 744B (40.8B active) Q8_0 776.4 GiB Does not fit even on 8 cards. Doesn't fit
Qwen3.8-2.4T-A95B 2.42T (93.3B active) Q8_0 2,524.1 GiB Does not fit even on 8 cards. Doesn't fit

The speed ceiling is memory bandwidth divided by the bytes read per generated token. It is a theoretical upper bound; public llama.cpp benchmarks typically reach about half to four-fifths of it, and small models and mixture-of-experts models often less. Public benchmarks

Detailed pages for this GPU

Other devices with 32 GB