| Model | Verdict | Decode tok/s | Max ctx | Size | Chat | Coding | |
|---|---|---|---|---|---|---|---|
| Qwen/Qwen3-30B-A3B-Instruct-2507 BF16 | Fits | 18172.6–200 | 256K | 30.5B | p85 | – | |
| Qwen/Qwen3.5-122B-A10B unsloth/Qwen3.5-122B-A10B-GGUF · Q4_K_M | Fits | 220187–253 | 256K | 125.1B | p77 | p49 | |
| zai-org/GLM-5.3-Flash peasantsmith/GLM-5.3-Flash-Maya-GGUF · IQ1_S | Fits | 24798.9–272 | 1M | 321.3B | p74 | p49 | |
| tencent/Hy3 bartowski/Hy3-GGUF · IQ1_M | Fits | 12048.2–132 | 219K | 298.8B | p73 | p46 | |
| Qwen/Qwen3.8-Flash-Next ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF · IQ3_S | Fits | 192163–221 | 256K | 180B | p68 | p51 | |
| google/gemma-4-31B-it unsloth/gemma-4-31B-it-GGUF · Q8_0 | Fits | 69.759.2–80.1 | 256K | 31.3B | p63 | – | |
| MiniMaxAI/MiniMax-M2.5 unsloth/MiniMax-M2.5-GGUF · UD-IQ2_M | Fits | 126108–145 | 192K | 228.7B | p46 | p61 | |
| Qwen/Qwen3.5-27B unsloth/Qwen3.5-27B-GGUF · Q8_0 | Fits | 75.964.5–87.3 | 256K | 27.8B | p60 | p46 | |
| Qwen/Qwen3.8-27B unsloth/Qwen3.8-27B-GGUF · Q8_0 | Fits | 75.163.9–86.4 | 256K | 27.8B | p59 | p33 | |
| thinkingmachines/Inkling-Small unsloth/Inkling-Small-GGUF · UD-IQ1_M | Fits | 213181–244 | 128K | 266B | p57 | p52 | |
| Qwen/Qwen3.6-27B unsloth/Qwen3.6-27B-GGUF · Q8_0 | Fits | 75.964.5–87.3 | 256K | 27.8B | p51 | p54 | |
| nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16 unsloth/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF · MXFP4_MOE | Fits | 148126–170 | 256K | 123.6B | p50 | p15 | |
| Qwen/Qwen3.5-35B-A3B BF16 | Fits | 18975.5–208 | 256K | 36B | p49 | p35 | |
| google/gemma-4-26B-A4B-it BF16 | Fits | 20682.2–226 | 256K | 25.8B | p49 | – | |
| openai/gpt-oss-120b unsloth/gpt-oss-120b-GGUF · F16 | Fits | 299254–343 | 128K | 117B | p47 | p15 | |
| Qwen/Qwen3.5-9B BF16 | Fits | 13152.2–144 | 256K | 9.7B | p46 | – | |
| ornith-ai/Ornith-1.5-9B BF16 | Fits | 13152.2–144 | 256K | 9.7B | p46 | p24 | |
| Qwen/Qwen3.6-35B-A3B BF16 | Fits | 18975.5–208 | 256K | 36B | p43 | p41 | |
| openbmb/MiniCPM-V-4.7-35B-A3B BF16 | Fits | 20783.0–228 | 256K | 35.2B | p42 | – | |
| ornith-ai/Ornith-1.0-35B BF16 | Fits | 21084.1–231 | 256K | 35.1B | – | p40 | |
| openai/gpt-oss-20b MXFP4 | Fits | 316126–347 | 128K | 21.1B | p39 | p15 | |
| deepseek-ai/DeepSeek-V4-Flash-0731 unsloth/DeepSeek-V4-Flash-0731-GGUF · UD-IQ1_S | Fits | 274233–315 | 1M | 282.1B | – | p39 | |
| Qwen/Qwen3.5-4B BF16 | Fits | 19678.3–215 | 256K | 4.7B | p39 | – | |
| google/gemma-4-12B-it BF16 | Fits | 10642.3–116 | 256K | 12B | p38 | – | |
| Qwen/Qwen3-4B-Instruct-2507 BF16 | Fits | 22590.0–248 | 256K | 4B | p38 | – | |
| nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 ggml-org/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-GGUF · BF16 | Fits | 194165–223 | 256K | 32.9B | p37 | p9 | |
| nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-NVFP4 MIXED_PRECISION | Fits | 257103–283 | 1M | 47.1B | p36 | p9 | |
| poolside/Laguna-S-2.1 Myric/Laguna-S-2.1-APEX-GGUF · V2 [quality] | Fits | 21586.0–236 | 1M | 117.6B | – | p34 | |
| nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 unsloth/Nemotron-3-Nano-30B-A3B-GGUF · BF16 | Fits | 194165–223 | 256K | 31.6B | p33 | – | |
| poolside/Laguna-XS-2.1 BF16 | Fits | 22489.5–246 | 256K | 33.4B | – | p30 | |
| Qwen/Qwen2.5-7B BF16 | Fits | 17569.9–192 | 128K | 7.6B | p29 | – | |
| google/gemma-4-E4B-it BF16 | Fits | 21184.5–232 | 128K | 8B | p27 | – | |
| openbmb/MiniCPM5-2B BF16 | Fits | 332133–366 | 128K | 2.5B | p27 | p3 | |
| poolside/Laguna-XS.2 BF16 | Fits | 22489.5–246 | 256K | 33.4B | – | p25 | |
| Qwen/Qwen3.5-2B BF16 | Fits | 335134–369 | 256K | 2.3B | p24 | – | |
| ornith-ai/Ornith-1.0-9B BF16 | Fits | 13353.3–147 | 256K | 9.4B | – | p20 | |
| google/gemma-4-E2B-it BF16 | Fits | 338135–372 | 128K | 5.1B | p18 | – | |
| Qwen/Qwen2.5-1.5B BF16 | Fits | 497199–547 | 128K | 1.5B | p15 | – | |
| mistralai/Mistral-7B-Instruct-v0.2 BF16 | Fits | 16164.3–177 | 32K | 7.2B | p14 | – | |
| Qwen/Qwen2.5-0.5B BF16 | Fits | 908363–999 | 32K | 0.5B | p5 | – | |
| Qwen/Qwen3.5-0.8B BF16 | Fits | 480192–527 | 256K | 0.9B | p4 | – | |
| incoai/Qwen3.8-27B-DFlash2 BF16 | Fits | 1736694–1909 | 256K | 1.9B | – | – | |
| Qwen/Qwen2.5-0.5B-Instruct BF16 | Fits | 908363–999 | 32K | 0.5B | – | – | |
| baidu/Unlimited-OCR-2 AtomicChat/Unlimited-OCR-2-GGUF · BF16 | Fits | 699594–804 | 128K | 3.3B | – | – | |
| Qwen/Qwen3-0.6B-Base BF16 | Fits | 553221–609 | 32K | 0.6B | – | – | |
| Qwen/Qwen3-0.6B BF16 | Fits | 524210–577 | 40K | 0.8B | – | – | |
| datalab-to/surya-ocr-2 BF16 | Fits | 509204–560 | 256K | 0.7B | – | – | |
| Qwen/Qwen2.5-1.5B-Instruct BF16 | Fits | 497199–547 | 32K | 1.5B | – | – | |
| Qwen/Qwen3.5-0.8B-Base BF16 | Fits | 480192–527 | 256K | 0.9B | – | – | |
| openbmb/MiniCPM-V-4.7-1B BF16 | Fits | 424170–466 | 256K | 1.6B | – | – | |
| Qwen/Qwen2-VL-2B-Instruct BF16 | Fits | 410164–451 | 32K | 2.2B | – | – | |
| baidu/Unlimited-OCR BF16 | Fits | 398159–438 | 32K | 3.3B | – | – | |
| Qwen/Qwen3-1.7B-Base BF16 | Fits | 395158–435 | 32K | 1.7B | – | – | |
| Qwen/Qwen3-1.7B BF16 | Fits | 366147–403 | 40K | 2B | – | – | |
| Qwen/Qwen3-VL-2B-Instruct BF16 | Fits | 358143–394 | 256K | 2.1B | – | – | |
| openbmb/MiniCPM5-2B-GGUF F16 | Fits | 332282–382 | 128K | 2.5B | – | – | |
| nvidia/Qwen3.6-35B-A3B-NVFP4 MIXED_PRECISION | Fits | 324129–356 | 256K | 34B | – | – | |
| ornith-ai/Ornith-1.5-35B-A3B-NVFP4 MIXED_PRECISION | Fits | 324129–356 | 256K | 34B | – | – | |
| Qwen/Qwen2.5-3B-Instruct BF16 | Fits | 301120–331 | 32K | 3.1B | – | – | |
| Qwen/Qwen2.5-VL-3B-Instruct BF16 | Fits | 267107–293 | 125K | 3.8B | – | – | |
| ornith-ai/Ornith-1.5-35B-A3B-FP8 COMPRESSED-TENSORS | Fits | 24698.4–271 | 256K | 36B | – | – | |
| Qwen/Qwen3-4B BF16 | Fits | 22590.0–248 | 40K | 4B | – | – | |
| HauhauCS/Gemma-4-E4B-Uncensored-HauhauCS-Aggressive Q8_K_P | Fits | 22590.0–247 | 128K | 7.5B | – | – | |
| ornith-ai/Ornith-1.0-35B-GGUF BF16 | Fits | 223190–257 | 256K | 34.7B | – | – | |
| Qwen/Qwen3-Coder-30B-A3B-Instruct-FP8 FP8 | Fits | 22087.8–241 | 256K | 30.5B | – | – | |
| ornith-ai/Ornith-1.5-35B-A3B-GGUF BF16 | Fits | 218185–251 | 256K | 35.5B | – | – | |
| ornith-ai/Ornith-1.5-9B-NVFP4 MIXED_PRECISION | Fits | 21686.4–238 | 256K | 9B | – | – | |
| Qwen/Qwen3-VL-4B-Instruct BF16 | Fits | 21284.9–234 | 256K | 4.4B | – | – | |
| nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16 unsloth/NVIDIA-Nemotron-3-Nano-4B-GGUF · BF16 | Fits | 204173–235 | 256K | 4B | – | – | |
| Qwen/Qwen3.5-4B-Base BF16 | Fits | 19678.3–215 | 256K | 4.7B | – | – | |
| ornith-ai/Ornith-1.5-122B-GGUF Q4_K_M | Fits | 195166–224 | 256K | 124.6B | – | – | |
| Qwen/Qwen3-Coder-30B-A3B-Instruct BF16 | Fits | 18172.6–200 | 256K | 30.5B | – | – | |
| Qwen/Qwen3-30B-A3B BF16 | Fits | 18172.6–200 | 40K | 30.5B | – | – | |
| datalab-to/chandra-ocr-2 BF16 | Fits | 18172.5–199 | 256K | 5.3B | – | – | |
| Qwen/Qwen2.5-7B-Instruct BF16 | Fits | 17569.9–192 | 32K | 7.6B | – | – | |
| Qwen/Qwen2.5-Coder-7B-Instruct BF16 | Fits | 17569.9–192 | 32K | 7.6B | – | – | |
| prism-ml/Bonsai-27B-mlx-1bit MLX-1BIT | Fits | 17268.6–189 | 256K | 26.4B | – | – | |
| Qwen/Qwen2.5-VL-7B-Instruct BF16 | Fits | 16365.0–179 | 125K | 8.3B | – | – | |
| Qwen/Qwen2-VL-7B-Instruct BF16 | Fits | 16365.0–179 | 32K | 8.3B | – | – | |
| Qwen/Qwen3-8B BF16 | Fits | 14959.5–164 | 40K | 8.2B | – | – | |
| Qwen/Qwen3-8B-Base BF16 | Fits | 14959.5–164 | 32K | 8.2B | – | – | |
| prism-ml/Ternary-Bonsai-27B-mlx-2bit MLX-2BIT | Fits | 14658.3–160 | 256K | 26.4B | – | – | |
| Qwen/Qwen3-VL-8B-Instruct BF16 | Fits | 14156.4–155 | 256K | 8.8B | – | – | |
| ornith-ai/Ornith-1.0-9B-GGUF BF16 | Fits | 137116–158 | 256K | 9B | – | – | |
| ornith-ai/Ornith-1.5-9B-GGUF BF16 | Fits | 133113–153 | 256K | 9.2B | – | – | |
| Qwen/Qwen2.5-Coder-14B-Instruct Qwen/Qwen2.5-Coder-14B-Instruct-GGUF · Q6_K | Fits | 10186.1–116 | 32K | 14.8B | – | – | |
| RadixArk/Qwen3.8-27B-NVFP4 MIXED_PRECISION | Fits | 91.236.5–100 | 256K | 26.4B | – | – | |
| Qwen/Qwen3-14B BF16 | Fits | 90.436.1–99.4 | 40K | 14.8B | – | – | |
| Qwen/Qwen2.5-14B-Instruct BF16 | Fits | 87.134.8–95.8 | 32K | 14.8B | – | – | |
| Qwen/Qwen3-32B unsloth/Qwen3-32B-GGUF · Q8_0 | Fits | 69.859.3–80.3 | 40K | 32.8B | – | – | |
| farbodtavakkoli/OTel-2.0-LLM-31B-IT wfakhri/OTel-2.0-LLM-31B-IT-GGUF · Q8_0 | Fits | 69.759.2–80.1 | 256K | 31.3B | – | – | |
| Qwen/Qwen2.5-Coder-32B-Instruct Qwen/Qwen2.5-Coder-32B-Instruct-GGUF · Q4_K_M | Fits | 63.453.9–72.9 | 32K | 32.8B | – | – | |
| Qwen/Qwen2.5-32B-Instruct BF16 | Fits | 41.916.8–46.1 | 32K | 32.8B | – | – | |
| Qwen/Qwen3.5-397B-A17B bartowski/Qwen_Qwen3.5-397B-A17B-GGUF · IQ2_M | Offload | 64.532.2–96.7 | 155K | 403.4B | p74 | p54 | |
| deepseek-ai/DeepSeek-V4-Flash antirez/deepseek-v4-gguf · V2 | Offload | 22690.4–339 | 1M | 282.1B | p62 | p61 | |
| poolside/Laguna-M.1 linuxid10t/Laguna-M.1-GGUF · Q4_K_M | Offload | 26.713.3–40.0 | 56K | 225.8B | – | p45 | |
| ornith-ai/Ornith-1.5-122B-FP8 COMPRESSED-TENSORS | Offload | 52.020.8–77.9 | 256K | 125.1B | – | – |
Quality = benchmark percentile against all indexed models (p50 = median). Speeds are single-user decode ranges.