Best AI models under $10 per million tokens
A leaderboard of the best AI models that cost $10 or less per million tokens. Claude Opus 5.5 is #1 right now, followed by GPT-6.1 Terra and GPT-6.1 Sol. Models are ranked by one score that combines 575 published benchmarks. Filter by your computer or budget to find the best one you can use.
Filter by your hardware and budget · up to $10.00/M · reset
Filter by task
New #1s:
orRSS
detecting your OS…
↗ share
$ terminal
Score vs release date
x
show
● open weights ◆ API only ─┘ best score so far · x: release month
| # | model | score↓ | where↓ | RAM↓ | speed↓ | sources |
|---|---|---|---|---|---|---|
| 2 |  Claude Opus 5.5 | 85±1 | api $8.00/M |       +1 | ||
| 3 |  GPT-6.1 Terra | 84±1 | api $2.00/M |  | ||
| 5 | GPT-6.1 Sol | 83±1 | api $4.00/M | | ||
| 8 | Claude Sonnet 5.5 | 82±1 | api $4.00/M | | ||
| 9 |  Gemini 4 Argon | 81±1 | api $4.00/M |  | ||
| 10 | Claude Opus 5 | 81±1 | api $10.00/M | +11 | ||
| 11 | GPT-6 Sol | 80±1 | api $4.00/M | +1 | ||
| 13 | GPT-5.6 Sol | 80±1 | api $8.00/M | +14 | ||
| 17 |  Muse Wonder | 78±1 | api $8.00/M |  | ||
| 19 |  MiMo-V2.6-Pro | 77±1 | api $0.54/M +local | ~443GB | | |
| 20 |  DeepSeek V4.1 Flash 1017 | 77±1 | api $0.26/M | | ||
| 21 | GPT-5.6 Terra | 77±1 | api $4.50/M | +5 | ||
| 22 |  Hy4 preview (770B-A49B) | 77±1 | api $1.25/M +local | ~248GB | | |
| 23 |  Qwen3.8 Max (2.4T-A95B) | 76±1 | api $3.00/M +local | ~380GB |  +2 | |
| 24 | Claude Opus 4.8 | 76±0 | api $10.00/M | +11 | ||
| 25 | GPT-5.4 | 76±1 | api $5.63/M | +14 | ||
| 26 | Gemini 3.8 Flash | 76±1 | api $1.50/M |  +5 | ||
| 27 | Muse Spark 1.3 | 76±1 | api $2.00/M | | ||
| 28 | MiMo-V2.6-Flash (309B-A15B) | 76±1 | api $0.17/M +local | ~136GB | | |
| 29 | DeepSeek V4.1 Flash | 76±1 | local | ~512GB | ~12 tok/s | |
| 31 |  Kimi K3 | 75±1 | api $6.00/M +local | ~884GB | +7 | |
| 32 | Claude Haiku 5.5 | 75±1 | api $0.20/M | | ||
| 33 | GPT-6 Luna | 75±1 | api $0.20/M | | ||
| 34 | GPT-5.6 Luna | 75±1 | api $0.45/M | +4 | ||
| 35 |  Grok 4.6 | 75±1 | api $3.00/M | | ||
| 36 | Gemini 3.7 Flash | 75±1 | api $1.50/M | +3 | ||
| 37 | DeepSeek V4 Pro (0813) | 75±1 | api $1.63/M +local | ~1010GB | +2 | |
| 38 |  GLM-5.3 | 75±1 | api $2.15/M +local | ~239GB | | |
| 39 | Qwen3.8 Max (0902) | 74±1 | api $3.00/M | | ||
| 40 | Muse Spark Open | 74±1 | api $2.00/M +local | ~190GB | | |
| 42 | Gemini 3.1 Pro (Preview) | 74±0 | api $4.50/M | +23 | ||
| 44 | Claude Opus 4.7 | 74±1 | api $10.00/M | +9 | ||
| 45 | Grok 4.5 | 74±1 | api $3.00/M | +1 | ||
| 46 | Qwen3.8-Omni-Flash | 74±1† | api $0.23/M | | ||
| 47 | DeepSeek V4 Flash Vision (Exp) | 74±2 | local | ~104GB | | |
| 48 | Qwen3.8-Flash-Next | 73±1 | local | ~59GB | ~27 tok/s | |
| 49 | Claude Sonnet 5 | 73±1 | api $4.00/M | +5 | ||
| 50 | Claude Opus 4.6 | 73±1 | api $10.00/M | +22 | ||
| 51 | Muse Spark 1.1 | 73±1 | api $2.00/M | | ||
| 52 | Grok 4.7 | 73±1 | api $3.00/M | | ||
| 53 | Muse Spark 1.2 | 73±1 | api $2.00/M | | ||
| 55 | Gemini 3.5 Flash | 73±1 | api $3.38/M | +3 | ||
| 56 | Qwen3.7 Max | 73±1 | api $3.75/M | | ||
| 57 | GPT-5.3-Codex | 73±1 | api $4.81/M | | ||
| 58 | DeepSeek V4 Flash (0731) | 72±1 | api $0.11/M +local | ~92GB | +2 | |
| 59 | GLM-5.3-Flash | 72±1 | api $0.24/M +local | ~103GB | +2 | |
| 60 | GPT-5.2 | 72±1 | api $4.81/M | +21 | ||
| 61 | Gemini 3.6 Flash | 72±1 | api $1.50/M | +2 | ||
| 62 | GLM-5.2 | 72±1 | api $2.15/M +local | ~121GB | +6 | |
| 63 | Grok 4.20 Multi-Agent (0309) | 72±2 | api $1.56/M | | ||
| 65 | Hunyuan Hy3 (295B-A21B) | 72±1 | api $0.23/M +local | ~94GB | | |
| 66 | Qwen3.8 27B | 72±1 | api $1.60/M +local | ~18GB | ~6.7 tok/s | +3 |
| 67 |  LongCat 2.0 (1.6T-A48B) | 72±2 | api $1.30/M +local | ~2155GB | | |
| 68 | Kimi K2.6 | 72±1 | api $1.71/M +local | ~317GB | +11 | |
| 71 | GPT-5 | 71±1 | api $3.44/M | +22 | ||
| 72 |  Mistral Large 4 (public preview) | 71±1 | api $2.06/M | | ||
| 73 | Qwen3.6 Max Preview | 71±1 | api $2.92/M | | ||
| 74 | Claude Sonnet 4.6 | 71±1 | api $6.00/M | +18 | ||
| 75 | Gemini 3 Flash (Preview) | 71±1 | api $1.13/M | +8 | ||
| 76 | DeepSeek V4 Pro (Preview, Apr 2026) | 71±1 | local | ~604GB | +6 | |
| 77 | Claude Opus 4.5 | 71±1 | api $10.00/M | +14 | ||
| 78 | Qwen3.7 Plus | 71±1 | api $0.70/M | | ||
| 79 | Kimi K2.7 Code | 71±1 | api $1.71/M +local | ~359GB | | |
| 81 |  Solar Open 2 (250B-A15B) | 71±1† | local | ~527GB | | |
| 82 | DeepSeek V4 Flash (Preview, Apr 2026) | 70±1 | local | ~103GB | +4 | |
| 83 | Grok 4.20 (0309 reasoning) | 70±1 | api $1.56/M | | ||
| 85 | GPT-5.4 mini | 70±1 | api $1.69/M | +6 | ||
| 86 |  Ling 3.0 Flash (124B-A5B) | 70±2† | api $0.090/M +local | ~71GB | | |
| 87 | GPT-5.1 | 70±1 | api $3.44/M | +8 | ||
| 88 | GLM-5.1 | 70±0 | api $2.15/M +local | ~119GB | +11 | |
| 90 | Qwen3.5 397B-A17B | 69±0 | api $1.09/M +local | ~65GB | +15 | |
| 92 |  KAT-Coder-V2.5-Dev (35B-A3B) | 69±2† | local | ~21GB | | |
| 93 |  Ornith 1.5 35B-A3B | 69±1† | local | ~21GB | ~80 tok/s | |
| 94 | Grok 4.3 | 69±1 | api $1.56/M | | ||
| 95 |  MiniMax-M3 | 69±1 | api $0.48/M +local | ~131GB | +2 | |
| 96 | Qwen3.6 Plus | 69±1 | api $1.13/M | +1 | ||
| 97 | Kimi K2.5 | 69±1 | api $1.20/M +local | ~160GB | +15 | |
| 100 | GLM-5 | 69±1 | api $1.55/M +local | ~119GB | +11 | |
| 101 | MiMo-V2.5-Pro (1T-A42B) | 69±1 | api $0.65/M +local | ~163GB | | |
| 102 | Muse Glimmer 30B | 69±1 | api $0.64/M +local | ~18GB | ~27 tok/s | |
| 103 | o3 | 68±1 | api $3.50/M | +11 | ||
| 104 |  Inkling Small (266B-A12B) | 68±1 | api $0.64/M +local | ~86GB | | |
| 105 |  Nemotron 3 Ultra 550B-A55B | 68±1 | api $1.35/M +local | ~89GB | | |
| 106 | Qwen3.6 27B | 68±1 | api $1.20/M +local | ~18GB | ~16 tok/s | +4 |
| 108 |  A.X K2 (688B-A33B) | 68±1† | local | ~1454GB | | |
| 109 | MiMo-V2.5 (310B-A15B) | 68±1 | api $0.21/M +local | ~100GB |  | |
| 110 | Qwen3.5 Plus | 68±2 | api $0.90/M | | ||
| 112 | Gemini 3.5 Flash-Lite | 68±1 | api $0.85/M | | ||
| 113 | Qwen3.5 122B-A10B | 68±1 | api $1.10/M +local | ~71GB | +5 | |
| 114 | Inkling (975B-A41B) | 68±1 | api $1.73/M +local | ~156GB | | |
| 115 | Claude Sonnet 4.5 | 68±1 | api $6.00/M | +19 | ||
| 116 | GLM-5V-Turbo | 68±2 | api $1.90/M | | ||
| 118 | Nemotron 3.5 Bolt 74B A9B | 68±2† | local | ~157GB | | |
| 119 | GPT-5.4 nano | 68±1 | api $0.46/M | +1 | ||
| 120 | GLM-4.7 | 68±1 | api $1.00/M +local | ~58GB | +5 | |
| 121 | Qwen3.5 27B | 68±0 | api $0.84/M +local | ~18GB | ~29 tok/s | +9 |
| 125 | Nemotron 3.5 Super VL 120B A12B | 67±1 | local | ~254GB | | |
| 126 | DeepSeek V3.2 | 67±1 | api $0.29/M +local | ~110GB | +15 | |
| 127 |  Step 3.5 Flash (196B-A11B) | 67±1 | api $0.15/M +local | ~118GB | ~18 tok/s | |
| 129 | Qwen3.6 35B-A3B | 67±1 | api $0.31/M +local | ~21GB | ~63 tok/s | +5 |
| 130 | Qwen3-Coder-Next (80B-A3B) | 67±2 | api $0.60/M +local | ~25GB | ~62 tok/s | |
| 131 | GPT-5 mini | 67±1 | api $0.69/M | +16 | ||
| 132 | MiniMax-M2.7 | 67±1 | api $0.53/M +local | ~75GB | +2 | |
| 133 | Ornith 1.0 35B (35B-A3B) | 67±1 | local | ~23GB | ~103 tok/s | |
| 135 | Gemini 2.5 Pro | 66±1 | api $3.44/M | +16 | ||
| 136 |  Gemma 4 31B | 66±1 | api $0.19/M +local | ~24GB | ~11 tok/s | +21 |
| 138 | Hy3 preview (295B-A21B) | 66±1 | local | ~629GB | | |
| 139 | Kimi K2 Thinking | 66±1 | api $1.07/M +local | ~160GB | | |
| 142 |  K-EXAONE 2.0 (750B-A37B) | 66±1† | local | ~791GB | | |
| 143 | Gemini 3.1 Flash-Lite | 66±1 | api $0.56/M | | ||
| 144 | Qwen3.7 Flash | 66±1 | api $0.055/M | | ||
| 145 |  Laguna M.1 | 66±2 | local | ~135GB | | |
| 146 | DeepSeek V3.1 Terminus | 66±1 | local | ~230GB | | |
| 147 | DeepSeek V3.2 Exp | 66±1 | local | ~726GB | | |
| 148 | Kimi K2 Instruct (0711) | 66±1 | api $1.00/M +local | ~152GB | +3 | |
| 150 |  GigaChat 3.5 Ultra Reasoning (432B-A28B) | 66±1† | local | ~461GB | | |
| 151 | o4-mini | 66±1 | api $1.93/M | +12 | ||
| 152 | Qwen3 235B-A22B Thinking 2507 | 66±1 | api $0.97/M +local | ~76GB | +6 | |
| 153 | Qwen3.5 35B-A3B | 66±0 | api $0.35/M +local | ~21GB | ~50 tok/s | +13 |
| 154 | Step 3.7 Flash (198B-A11B) | 66±1 | api $0.44/M +local | ~104GB | | |
| 155 | Grok 4.1 Fast (Reasoning) | 66±1 | api $0.28/M | +4 | ||
| 156 | LongCat-Flash-Chat (560B-A27B) | 66±1 | local | ~333GB | | |
| 157 | Qwen3.6 Flash | 65±1 | api $0.56/M | | ||
| 158 |  Spark-X2.5 4B | 65±1† | local | ~3.5GB | ~114 tok/s | |
| 159 | DeepSeek V3.1 | 65±1 | api $0.42/M +local | ~110GB | +3 | |
| 160 | GLM-4.6 | 65±1 | api $1.00/M +local | ~58GB | +1 | |
| 162 | Qwen3-Max (2025-09-23) | 65±1 | api $2.40/M | | ||
| 163 | MiniMax-M2.5 | 65±1 | api $0.53/M +local | ~75GB | +3 | |
| 164 | o3-mini | 65±1 | api $1.93/M | +13 | ||
| 165 | MiMo-V2-Flash (309B-A15B) | 65±1 | local | ~103GB | | |
| 166 | DeepSeek R1 0528 | 65±1 | api $0.91/M +local | ~104GB | +6 | |
| 168 | MiniMax-M2.1 | 65±1 | local | ~79GB | | |
| 170 | Nemotron Cascade 2 30B-A3B | 65±1 | local | ~20GB | ~90 tok/s | |
| 171 | MiMo-V2.6-Distill-Qwen-9B | 64±2† | local | ~7.8GB | | |
| 172 | Laguna XS 2.1 (33B-A3B) | 64±2† | local | ~21GB | | |
| 173 | K-EXAONE (236B-A23B) | 64±1 | local | ~152GB | | |
| 174 | Qwen3 235B-A22B Instruct 2507 | 64±1 | api $0.20/M +local | ~73GB | ~15 tok/s | +3 |
| 175 | Mistral Medium 3.5 (128B) | 64±1 | api $3.00/M +local | ~43GB | | |
| 176 |  INTELLECT-3 | 64±1 | local | ~48GB | | |
| 177 | GLM-4.5 | 64±1 | api $1.00/M +local | ~58GB | +3 | |
| 178 | Qwen3.5 Flash | 64±1 | api $0.17/M | | ||
| 179 | MiniMax-M2 | 64±1 | api $0.53/M +local | ~75GB | | |
| 181 |  Command A+ (05-2026) | 64±1 | local | ~71GB |  | |
| 182 | Claude Haiku 4.5 | 64±1 | api $2.00/M | +9 | ||
| 183 | Nemotron 3 Super 120B-A12B | 64±1 | api $0.28/M +local | ~73GB | ~20 tok/s | |
| 184 | Laguna XS.2 | 64±1 | local | ~21GB | | |
| 185 | Gemma 4 26B-A4B | 64±1 | api $0.14/M +local | ~18GB | ~55 tok/s | +13 |
| 186 | Solar Pro 4 | 64±1 | api $0.53/M | | ||
| 187 | EXAONE 4.5 33B | 64±2† | local | ~24GB | | |
| 188 | Kimi K2 Instruct (0905) | 63±2 | local | ~298GB | | |
| 189 | Qwen3-Coder 480B-A35B Instruct | 63±1 | api $0.67/M +local | ~78GB | | |
| 191 | Qwen3-VL 235B-A22B Thinking | 63±1 | local | ~106GB | | |
| 192 | DeepSeek R1 (Jan '25) | 63±1 | api $4.00/M +local | ~104GB | +15 | |
| 195 | Qwen3-VL 235B-A22B Instruct | 63±1 | local | ~128GB | | |
| 196 |  MiniCPM5 2B | 63±1† | local | ~2.4GB | | |
| 197 | Nemotron-Labs-3 Puzzle 75B-A9B | 63±1† | local | ~35GB | | |
| 198 | Qwen3-Next 80B-A3B Instruct | 63±1 | api $0.34/M +local | ~26GB | ~55 tok/s | |
| 199 |  Nova 2 Pro (Preview) | 63±2 | api $3.78/M | | ||
| 201 | Qwen3-Next 80B-A3B Thinking | 63±1 | api $0.41/M +local | ~25GB | +3 | |
| 202 |  Granite 4.2 30B | 63±1 | api $0.28/M +local | ~19GB | | |
| 205 | GPT-4.1 | 62±1 | api $3.50/M | +17 | ||
| 206 | GLM-4.5-Air | 62±1 | api $0.42/M +local | ~35GB | ~23 tok/s | +4 |
| 207 |  Trinity-Large-Thinking | 62±2 | api $0.39/M +local | ~141GB | | |
| 208 | Qwen3.5 9B | 62±0 | api $0.11/M +local | ~6.7GB | ~22 tok/s | +19 |
| 209 | Nemotron 3.5 Lightning 30B-A3B | 62±2 | api $0.087/M +local | ~20GB | ~64 tok/s | |
| 210 |  Seed-OSS 36B Instruct | 62±1 | local | ~23GB | | |
| 211 | Gemma 4 12B | 62±1 | local | ~9.2GB | ~29 tok/s | |
| 213 | MiniMax-M1 (80k) | 62±1 | api $0.96/M +local | ~959GB | | |
| 214 | GLM-4.6V | 62±2 | api $0.45/M +local | ~50GB | | |
| 216 | Gemini 2.5 Flash | 62±1 | api $0.85/M | +10 | ||
| 219 | Qwen3-VL 32B Thinking | 62±1 | api $0.28/M +local | ~22GB | | |
| 220 | gpt-oss-120b | 62±1 | api $0.26/M +local | ~64GB | ~56 tok/s | +19 |
| 221 | GPT-5 nano | 62±1 | api $0.14/M | +7 | ||
| 222 | Mistral Large 3 (675B-A41B) | 62±1 | api $0.75/M +local | ~108GB | +1 | |
| 223 |  Apriel 1.6 15B Thinker | 62±1† | local | ~10GB | | |
| 224 | Devstral 2 (2512) | 62±2 | local | ~74GB | | |
| 226 | A.X K1 (519B-A33B) | 61±1† | local | ~1091GB | | |
| 228 | GLM-4.5V | 61±2 | api $0.90/M +local | ~69GB | | |
| 229 | Devstral Small 2 24B | 61±2 | local | ~16GB | | |
| 231 | Llama 3.1 Nemotron Ultra 253B v1 | 61±1 | local | ~82GB | | |
| 234 | DeepSeek V3 0324 | 60±1 | local | ~102GB | +11 | |
| 237 | GPT-4.1 mini | 60±1 | api $0.70/M | +13 | ||
| 238 | Qwen3 30B-A3B Instruct 2507 | 60±1 | api $0.15/M +local | ~19GB | ~100 tok/s | +4 |
| 239 | Qwen3 30B-A3B Thinking 2507 | 60±1 | api $0.75/M +local | ~19GB | +8 | |
| 240 |  ERNIE 4.5 300B-A47B | 60±1† | local | ~97GB | | |
| 241 | Qwen3.5 4B | 60±1 | local | ~3.8GB | +11 | |
| 243 |  Mellum2.1 12B-A2.5B Thinking | 60±1† | local | ~8.3GB | | |
| 244 |  Gemma 2 9B IT SimPO | 60±1 | local | ~7.6GB | | |
| 245 | Gemma 2 9B IT DPO (Princeton) | 60±1 | local | ~7.6GB | | |
| 246 |  Falcon-H1R 7B | 60±1† | local | ~5.3GB | | |
| 247 | Nemotron-Cascade-14B-Thinking | 60±1† | local | ~10GB | | |
| 248 | Qwen3-VL 32B Instruct | 60±1 | api $0.28/M +local | ~22GB | | |
| 249 | DeepSeek V3 (Dec '24) | 60±1 | api $0.46/M +local | ~337GB | +10 | |
| 250 | Qwen3 235B-A22B (Apr 2025, hybrid) | 60±0 | api $1.23/M +local | ~73GB | +14 | |
| 252 | Mistral Small 4 (119B-A6.5B) | 60±2 | api $0.26/M +local | ~37GB | | |
| 253 | Granite 4.2 8B | 59±1 | api $0.11/M +local | ~6.8GB | | |
| 254 | QwQ-32B | 59±1 | local | ~21GB | +7 | |
| 255 | Solar Open 100B (102B-A12B) | 59±1† | local | ~41GB | | |
| 256 | Qwen3-VL 30B-A3B Thinking | 59±1 | api $0.75/M +local | ~20GB | | |
| 257 | Trinity-Large-Preview | 59±2 | local | ~155GB | | |
| 258 | OpenReasoning-Nemotron-32B | 59±2† | local | ~21GB | | |
| 259 | EXAONE 4.0 32B | 59±1 | local | ~20GB | | |
| 260 |  Phi-4-reasoning-plus 14B | 59±1 | local | ~10GB | | |
| 261 | GLM-4.7-Flash | 58±1 | api $0.0000/M +local | ~18GB | ~44 tok/s | +1 |
| 263 | gpt-oss-20b | 58±1 | api $0.13/M +local | ~13GB | ~79 tok/s | +16 |
| 264 |  Olmo 3.1 32B Think | 58±1 | local | ~21GB | | |
| 266 | Qwen3 14B | 58±1 | api $0.61/M +local | ~10GB | +7 | |
| 267 | Llama 3.3 Nemotron Super 49B v1 | 58±2 | local | ~22GB | | |
| 268 | DeepSeek R1 Distill Llama 70B | 58±1 | api $0.80/M +local | ~30GB | +1 | |
| 269 | GPT-4o (2024-08-06) | 58±1 | api $4.38/M | +23 | ||
| 270 | Nemotron-Cascade-8B | 58±1† | local | ~6.3GB | | |
| 271 | Command A (03-2025) | 58±1 | api $4.38/M +local | ~37GB | | |
| 272 | Olmo 3 32B Think | 58±1 | local | ~21GB | | |
| 273 | GPT-4o (2024-05-13) | 58±0 | api $7.50/M | +19 | ||
| 274 | Phi-4-reasoning 14B | 58±1 | api $0.22/M +local | ~10GB | | |
| 276 | Nova 2 Lite | 58±1 | api $0.94/M | | ||
| 277 | Magistral Small 1.2 | 58±1 | api $0.75/M +local | ~16GB | | |
| 278 | GPT-4o (2024-11-20) | 58±1 | api $4.38/M | +7 | ||
| 279 | Qwen3-VL 8B Thinking | 58±1 | api $0.66/M +local | ~6.9GB | | |
| 280 | Qwen3 32B | 58±0 | api $0.13/M +local | ~20GB | +17 | |
| 283 | Qwen3-VL 30B-A3B Instruct | 58±1 | api $0.35/M +local | ~20GB | | |
| 284 | Qwen3 Coder 30B-A3B Instruct | 57±1 | api $0.12/M +local | ~19GB | ~93 tok/s | |
| 285 | Qwen2.5 Coder 32B Instruct | 57±1 | api $0.74/M +local | ~21GB | +5 | |
| 286 | Gemini 2.5 Flash-Lite | 57±2 | api $0.17/M | +1 | ||
| 287 | Llama 4 Maverick (402B-A17B) | 57±1 | api $0.42/M +local | ~65GB | +14 | |
| 288 | DeepSeek R1 0528 Qwen3 8B | 57±1 | api $0.068/M +local | ~6.5GB | ~21 tok/s | |
| 290 | Granite 4.2 3B | 57±1 | api $0.052/M +local | ~3.3GB | | |
| 292 | Llama 3.3 Nemotron Super 49B v1.5 | 57±1 | local | ~18GB | | |
| 293 | Nemotron 3 Nano 30B-A3B | 57±1 | api $0.087/M +local | ~20GB | ~90 tok/s | +3 |
| 294 |  Athene-V2-Chat 72B | 57±1 | local | ~44GB | | |
| 296 | Granite 4.1 30B | 57±1† | local | ~19GB | ~25 tok/s | |
| 297 | Nemotron-Cascade-8B-Thinking | 57±2 | local | ~6.3GB | | |
| 299 | Qwen2.5-Omni 7B | 57±2 | local | ~5.8GB | | |
| 300 |  Sky-T1 32B Preview | 57±4 | local | ~21GB | | |
| 301 | Ministral 3 14B Reasoning 2512 | 57±1 | local | ~10GB | | |
| 302 | Qwen Plus (2025-01-25) | 57±1 | api $0.60/M | | ||
| 303 | Mistral Small 3.2 24B | 56±1 | api $0.11/M +local | ~15GB | +4 | |
| 304 | DeepSeek-V2.5 (1210) | 56±1 | local | ~82GB | | |
| 305 |  QED-Nano | 56±2 | local | ~4.1GB | | |
| 306 | Qwen3 4B Thinking 2507 | 56±1 | local | ~4.2GB | ~67 tok/s | +3 |
| 310 | Qwen3-VL 4B Thinking | 56±1 | local | ~4.4GB | | |
| 311 | Qwen2.5 72B Instruct | 56±1 | api $0.37/M +local | ~31GB | +24 | |
| 312 | Qwen2.5-VL 72B Instruct | 56±1 | local | ~46GB | +3 | |
| 313 | Llama 3.1 405B Instruct | 56±1 | api $2.40/M +local | ~160GB | +13 | |
| 314 | Mistral Large 2 (2411) | 56±1 | local | ~77GB | +8 | |
| 315 | Qwen2.5 Coder 14B Instruct | 56±1 | local | ~11GB | ~18 tok/s | |
| 316 | DeepSeek R1 Distill Qwen 32B | 56±1 | api $1.59/M +local | ~21GB | ~7.3 tok/s | +5 |
| 317 | Qwen2.5 32B Instruct | 56±1 | local | ~21GB | +11 | |
| 318 | Qwen3 8B | 56±1 | api $0.31/M +local | ~6.3GB | ~25 tok/s | +15 |
| 319 | DeepSeek R1 Distill Qwen 14B | 56±1 | local | ~11GB | ~15 tok/s | +2 |
| 321 | Molmo 2 8B | 56±2 | local | ~6.8GB | | |
| 322 | Ministral 3 14B | 55±1 | api $0.20/M +local | ~9.1GB | ~46 tok/s | +2 |
| 323 | Nemotron Nano 9B v2 | 55±1 | api $0.10/M +local | ~7.6GB | | |
| 324 | EXAONE Deep 32B | 55±2† | local | ~22GB | | |
| 325 | Qwen3 4B Instruct 2507 | 55±1 | local | ~4.2GB | ~107 tok/s | +4 |
| 326 | Qwen3 30B-A3B (Apr 2025, hybrid) | 55±0 | api $0.35/M +local | ~18GB | +10 | |
| 327 |  Bielik 11B v3.0 Instruct | 55±2 | local | ~8.1GB | | |
| 328 | Nova Premier | 55±2† | api $5.00/M |  | ||
| 329 | Granite 4.1 8B | 55±1 | api $0.11/M +local | ~7.1GB | ~70 tok/s | |
| 330 | Magistral Small 1.0 (2506) | 55±1 | local | ~16GB | | |
| 331 | Spark-X2.5 1.7B | 55±1† | local | ~1.9GB | ~166 tok/s | |
| 332 | Qwen2.5-Math 72B Instruct | 55±2† | local | ~44GB | | |
| 333 | Llama 3.3 70B Instruct | 55±1 | api $0.72/M +local | ~25GB | ~13 tok/s | +21 |
| 335 | Qwen3-VL 8B Instruct | 55±1 | api $0.31/M +local | ~6.9GB | | |
| 336 | Mistral Large 2 (2407) | 55±1 | local | ~45GB | +10 | |
| 337 | Falcon-H1 34B Instruct | 55±3 | local | ~23GB | | |
| 338 | Llama 4 Scout (109B-A17B) | 55±1 | api $0.15/M +local | ~36GB | ~19 tok/s | +8 |
| 340 | Bielik 11B v2.6 Instruct | 55±3 | local | ~8.1GB | | |
| 341 | Mistral Small 3.1 24B (2503) | 55±1 | local | ~16GB | +7 | |
| 342 | Llama 3.1 Tulu 3 70B | 55±1 | local | ~43GB | | |
| 343 | Pixtral Large (2411) | 55±3 | local | ~51GB | | |
| 344 | Olmo 3.1 32B Instruct | 55±2 | local | ~21GB | | |
| 345 | GigaChat 3.5 Ultra Instruct (432B-A28B) | 55±1† | local | ~457GB | | |
| 347 | DeepSeek-V2.5 | 55±1 | local | ~151GB | +1 | |
| 348 |  OpenThinker2 32B | 55±1 | local | ~21GB | | |
| 349 | DeepSeek-V2 Chat (0628) | 55±1 | local | ~66GB | | |
| 351 | OpenThinker 32B | 55±1† | local | ~21GB | | |
| 352 | Command A Reasoning (08-2025) | 54±2 | local | ~34GB | | |
| 353 | Magistral Small 1.1 (2507) | 54±1† | local | ~16GB | | |
| 354 | Mellum2 12B-A2.5B Thinking | 54±1† | local | ~8.6GB | | |
| 355 | Ministral 3 8B Reasoning 2512 | 54±1 | local | ~6.5GB | | |
| 356 | Llama 3.1 Tulu 3 70B DPO | 54±1 | local | ~22GB | | |
| 358 | Nemotron 3 Nano 4B | 54±1 | local | ~2.9GB | ~52 tok/s | |
| 359 | GPT-4o mini (2024-07-18) | 54±0 | api $0.26/M | +32 | ||
| 360 | DeepSeek-Coder-V2 Instruct (0614) | 54±1 | local | ~92GB | +2 | |
| 362 | DeepSeek-Coder-V2 Instruct (0724) | 54±1 | local | ~151GB | | |
| 363 | Mistral Small 3 24B (2501) | 54±1 | local | ~15GB | +6 | |
| 367 |  GLM-4 9B Chat 1M | 54±4 | local | ~8.1GB | | |
| 368 | EXAONE Deep 7.8B | 54±2† | local | ~6.5GB | | |
| 371 | Qwen2-VL 72B Instruct | 54±1 | local | ~34GB | | |
| 372 |  Hermes 3 Llama 3.1 405B | 54±1 | local | ~114GB | | |
| 373 | Llama 3.1 70B Instruct | 53±0 | api $0.72/M +local | ~27GB | +22 | |
| 374 | Gemma 3 27B | 53±1 | api $0.10/M +local | ~19GB | +23 | |
| 375 | Qwen2.5-VL 32B Instruct | 53±2 | local | ~23GB | +1 | |
| 376 | GPT-4.1 nano | 53±1 | api $0.17/M | +11 | ||
| 378 | Phi-4 14B | 53±1 | api $0.087/M +local | ~10GB | +11 | |
| 379 | Bielik 11B v2.1 Instruct | 53±3 | local | ~8.5GB | | |
| 380 | Qwen2.5-Math 7B Instruct | 53±2† | local | ~5.4GB | | |
| 381 | Olmo 3 7B Think | 53±1 | local | ~5.8GB | | |
| 382 |  s1.1 32B | 53±1 | local | ~22GB | | |
| 383 | Ministral 3 8B | 53±1 | api $0.15/M +local | ~6.5GB | ~68 tok/s | +3 |
| 384 | Gemma 4 E4B | 53±1 | local | ~6.4GB | ~21 tok/s | +8 |
| 385 | Llama 3.1 Nemotron 70B Instruct | 53±1 | local | ~43GB | +4 | |
| 386 | Qwen3-VL 4B Instruct | 53±1 | local | ~4.4GB | | |
| 390 |  LFM2.5 2.6B | 53±1 | local | ~2.3GB | | |
| 391 | Bielik 11B v2.2 Instruct | 53±3 | local | ~8.5GB | | |
| 392 | Llama 3.1 Nemotron 51B Instruct | 53±2 | local | ~34GB | | |
| 393 | Aya Expanse 32B | 53±2 | api $0.75/M +local | ~22GB | +1 | |
| 395 | Granite 4.0 H Small (32B-A9B) | 53±1 | api $0.11/M +local | ~21GB | +1 | |
| 396 | Qwen2.5 Coder 7B Instruct | 53±1 | local | ~5.4GB | ~3.8 tok/s | |
| 397 |  OpenCoder-8B-Instruct | 53±1 | local | ~6.4GB | | |
| 398 | LFM2.5 8B-A1B | 52±1 | local | ~5.5GB | ~174 tok/s | |
| 399 | Gemma 3 12B | 52±1 | api $0.075/M +local | ~10GB | ~13 tok/s | +10 |
| 400 | Qwen2.5 14B Instruct | 52±1 | local | ~11GB | +7 | |
| 402 |  Cogito v1 Preview Llama 8B | 52±4 | local | ~6.2GB | | |
| 403 | Llama-3-Instruct-8B-SimPO-ExPO | 52±1 | local | ~6.6GB | | |
| 404 | Llama-3-Instruct-8B-SimPO | 52±1 | local | ~6.6GB | | |
| 405 | Athene-70B | 52±1 | local | ~29GB | | |
| 406 | Qwen2 72B Instruct | 52±1 | local | ~33GB | +14 | |
| 407 |  Jamba 1.5 Large | 52±1 | local | ~839GB | | |
| 409 |  CodeQwen1.5 7B Chat | 52±1 | local | ~6.4GB | | |
| 410 |  OpenCodeInterpreter-DS-33B | 51±1 | local | ~22GB | | |
| 411 | Bielik 11B v2.0 Instruct | 51±3 | local | ~8.5GB | | |
| 414 | LFM2 24B-A2B | 51±1 | local | ~14GB | | |
| 415 | Qwen3 4B | 51±1 | local | ~4GB | +8 | |
| 416 | DeepSeek-Coder 33B Instruct | 51±1 | local | ~21GB | +1 | |
| 418 | Codestral 22B v0.1 | 51±1 | local | ~15GB | | |
| 419 | Gemma 2 27B IT | 51±1 | local | ~19GB | +10 | |
| 420 | Llama 3 70B Instruct | 51±0 | local | ~24GB | +17 | |
| 421 | Llama 3.2 90B Vision Instruct | 51±1 | local | ~58GB | +1 | |
| 422 |  WizardCoder 33B V1.1 | 51±1 | local | ~22GB | | |
| 423 | Bielik 11B v2.3 Instruct | 51±2 | local | ~8.6GB | | |
| 425 | Llama-3-8B-ProLong-512k-Instruct | 51±4 | local | ~35GB | | |
| 426 | Nova Pro | 51±1 | api $1.40/M | | ||
| 427 | OpenCodeInterpreter-DS-6.7B | 51±1 | local | ~5.8GB | | |
| 428 | Qwen2-Math 72B Instruct | 51±1 | local | ~32GB | | |
| 429 | WizardLM-2 8x22B | 51±1 | local | ~91GB | | |
| 430 | Mistral Small Instruct 22B (2409) | 51±2 | local | ~15GB | | |
| 431 | OpenThinker3 7B | 51±1 | local | ~6GB | | |
| 432 | Granite 4.1 3B | 51±1† | local | ~3.7GB | ~131 tok/s | |
| 433 |  Artigenz-Coder-DS-6.7B | 51±1 | local | ~5.8GB | | |
| 434 |  Yi 34B 200K (base) | 51±4 | local | ~23GB | | |
| 435 | Mellum2 12B-A2.5B Instruct | 51±2† | local | ~7.6GB | | |
| 436 | Molmo 72B (0924) | 50±2 | local | ~309GB | | |
| 437 | Hermes 3 Llama 3.1 70B | 50±1 | local | ~43GB | | |
| 438 |  speechless-codellama-34b-v2.0 | 50±1 | local | ~23GB | | |
| 439 | OLMo 2 32B Instruct (0325) | 50±1 | local | ~21GB | | |
| 440 | Ministral 3 3B Reasoning 2512 | 50±1 | local | ~3.4GB | | |
| 441 |  Aya 23 35B | 50±2 | local | ~22GB | | |
| 442 | AceReason-Nemotron-1.0-14B | 50±2† | local | ~10GB | | |
| 444 | Qwen2.5-VL 7B Instruct | 50±1 | local | ~5.8GB | | |
| 445 | Mixtral 8x22B Instruct v0.1 | 50±1 | local | ~56GB | +9 | |
| 446 |  Magicoder-S-DS 6.7B | 50±1 | local | ~5.8GB | | |
| 447 | EXAONE Deep 2.4B | 50±2† | local | ~3.2GB | | |
| 448 | Pixtral 12B (2409) | 50±2 | local | ~9.6GB | | |
| 449 | Nova Lite | 50±2 | api $0.10/M | | ||
| 450 |  XwinCoder-34B | 50±1 | local | ~23GB | | |
| 451 | Qwen2.5 7B Instruct | 50±1 | api $0.30/M +local | ~5.4GB | ~54 tok/s | +12 |
| 452 | LFM2.5 1.2B Thinking | 50±2 | local | ~1.3GB | | |
| 453 | Yi-Coder 9B Chat | 50±1 | local | ~6.3GB | | |
| 454 |  InternLM2.5 7B Chat | 50±2 | local | ~6.1GB | | |
| 455 |  StarChat2 15B v0.1 | 50±1 | local | ~12GB | | |
| 456 |  Code Millenials 34B | 50±1 | local | ~23GB | | |
| 457 | Qwen2-Math 7B Instruct | 50±2† | local | ~6GB | | |
| 458 | WaveCoder-Ultra-6.7B | 50±1 | local | ~5.8GB | | |
| 461 |  MegaBeam-Mistral-7B-512k | 49±4 | local | ~5.9GB | | |
| 463 | Phi-3.5-MoE-instruct | 49±1 | local | ~25GB | | |
| 464 | Qwen3-VL 2B Thinking | 49±1 | local | ~2.7GB | | |
| 465 | Gemma 3n E4B | 49±1 | local | ~5.6GB | | |
| 467 | DeepSeek-Coder 7B Instruct v1.5 | 49±1 | local | ~5.9GB | | |
| 468 | OpenThinker2 7B | 49±2† | local | ~6.4GB | | |
| 469 | DeepSeek-Coder 6.7B Instruct | 49±1 | local | ~8.8GB | | |
| 470 | Qwen2 57B-A14B Instruct | 49±2 | local | ~35GB | | |
| 471 | Hermes 2 Theta Llama-3 70B | 49±2 | local | ~47GB | | |
| 472 | MiniCPM3-4B | 49±2 | local | ~4GB | | |
| 473 | Granite 4.0 H Tiny (7B-A1B) | 49±1 | local | ~4.5GB | ~12 tok/s | |
| 474 | Qwen2.5 Coder 3B Instruct | 49±1 | local | ~2.6GB | ~9.5 tok/s | |
| 476 | Gemma 2 9B IT | 49±1 | local | ~8.9GB | +14 | |
| 477 | Yi-1.5 34B (base) | 49±2 | local | ~22GB | | |
| 478 | Bielik 4.5B v3.0 Instruct | 49±2 | local | ~4.3GB | | |
| 480 |  WhiteRabbitNeo-33B-v1 | 49±1 | local | ~22GB | | |
| 481 | Qwen1.5 110B Chat | 48±1 | local | ~45GB | +1 | |
| 482 | Command R Plus (08-2024) | 48±1 | api $4.38/M +local | ~63GB | +1 | |
| 483 | GLM-4 9B Chat | 48±2 | local | ~7.1GB | +1 | |
| 484 | speechless-coder-ds-6.7b | 48±1 | local | ~5.8GB | | |
| 485 | Magicoder-S-CL-7B | 48±1 | local | ~5.8GB | | |
| 486 |  Phind CodeLlama 34B v2 | 48±1 | local | ~23GB | | |
| 487 | AceReason-Nemotron-1.1-7B | 48±2† | local | ~6GB | | |
| 488 | Jamba 1.5 Mini | 48±1 | local | ~34GB | | |
| 489 | DeepSeek R1 Distill Qwen 7B | 48±1 | local | ~5.4GB | ~35 tok/s | +7 |
| 492 | GPT-3.5 Turbo (0125) | 48±1 | api $0.75/M | +11 | ||
| 493 |  InternVL2-26B | 48±2 | local | ~55GB | | |
| 494 | Aya Expanse 8B | 48±2 | local | ~6.4GB | | |
| 495 |  EuroLLM 9B Instruct | 48±2 | local | ~6.9GB | | |
| 498 | Granite 4.0 H Micro 3B | 48±1 | local | ~3.4GB | | |
| 500 | Code Llama 70B Instruct | 48±1 | local | ~28GB | +1 | |
| 501 | Command R (08-2024) | 48±1 | api $0.26/M +local | ~21GB | +3 | |
| 502 | OLMo 2 13B Instruct (1124) | 48±1 | local | ~9.6GB | | |
| 503 | InternLM2.5 7B Chat 1M | 48±4 | local | ~6.4GB | | |
| 504 | Falcon-H1 7B Instruct | 48±3 | local | ~6GB | | |
| 506 | Llama 3 70B (base) | 47±2 | local | ~43GB | | |
| 507 | Nova Micro | 47±2 | api $0.061/M | | ||
| 508 | Qwen3.5 2B | 47±1 | local | ~2.2GB | ~137 tok/s | +7 |
| 509 | Tulu 2 70B | 47±2† | local | ~28GB | | |
| 511 | Hermes 2 Pro Llama 3 8B | 47±3 | local | ~6.6GB | | |
| 512 |  MedGemma 4B IT | 47±2 | local | ~4.1GB | | |
| 514 | Yi 1.5 34B Chat | 47±1 | local | ~22GB | +5 | |
| 515 |  OpenChat 3.5 0106 | 47±1 | local | ~6.1GB | +1 | |
| 517 | WizardCoder Python 34B V1.0 | 47±1 | local | ~22GB | | |
| 519 | Granite 4.0 Micro 3B | 47±1 | local | ~3.5GB | ~6.2 tok/s | |
| 520 | Gemma 4 E2B | 47±1 | api $0.050/M +local | ~4.5GB | +6 | |
| 522 |  DBRX Instruct | 47±1 | local | ~85GB | +3 | |
| 524 | Granite 34B Code Instruct (8K) | 47±1 | local | ~22GB | | |
| 525 | Cogito v1 Preview Llama 3B | 47±4 | local | ~3.4GB | | |
| 526 | Qwen2.5-Math 1.5B Instruct | 47±2† | local | ~1.8GB | | |
| 527 | EXAONE 4.0 1.2B | 47±1† | local | ~2.2GB | | |
| 528 | InternLM2.5 20B Chat | 47±2 | local | ~14GB | | |
| 529 |  StarCoder2-15B-Instruct-v0.1 | 47±1 | local | ~11GB | | |
| 531 | Qwen1.5 72B Chat | 47±1 | local | ~31GB | +9 | |
| 532 | Granite 3.3 8B Instruct | 47±1 | local | ~6.3GB | | |
| 533 | speechless-starcoder2-15b | 47±1 | local | ~12GB | | |
| 534 | OpenChat 3.5 0106 Gemma | 47±2 | local | ~6.9GB | | |
| 535 | Phi-3 Medium 128K Instruct | 47±1 | local | ~10GB | | |
| 536 | Zephyr ORPO 141B-A35B v0.1 | 47±2 | local | ~56GB | | |
| 537 | Qwen2-VL 7B Instruct | 47±1 | local | ~5.8GB | | |
| 541 | DeepSeek R1 Distill Llama 8B | 46±1 | local | ~6.2GB | +2 | |
| 542 | Phi-3-medium-4k-instruct | 46±1 | local | ~10GB | +1 | |
| 543 | QwQ-32B-Preview | 46±2 | local | ~21GB | +2 | |
| 544 | DeepSeek-Coder-V2 Lite Instruct (16B-A2.4B) | 46±1 | local | ~9.6GB | +1 | |
| 546 | Qwen2-Math 1.5B Instruct | 46±2† | local | ~2.4GB | | |
| 548 | Llama 3.1 8B Instruct | 46±0 | api $0.22/M +local | ~6.2GB | ~47 tok/s | +25 |
| 549 | Phi-4-mini-reasoning 3.8B | 46±2 | api $0.13/M +local | ~4GB | | |
| 550 | Ministral 3 3B | 46±1 | api $0.10/M +local | ~3.4GB | ~128 tok/s | +2 |
| 551 | Olmo 3 7B Instruct | 46±1 | local | ~5.8GB | | |
| 553 | Command R Plus (04-2024) | 46±1 | local | ~43GB | +6 | |
| 554 | Mistral NeMo Instruct (2407) | 46±1 | local | ~8.8GB | +9 | |
| 556 | Llama 3.1 Tulu 3 8B | 46±1 | local | ~6.2GB | | |
| 557 | Phi-3-mini-4k-instruct (June 2024 update) | 46±1 | local | ~5.7GB | | |
| 558 | Yi 1.5 9B Chat | 46±1 | local | ~6.8GB | +1 | |
| 559 | Qwen1.5 32B Chat | 46±1 | local | ~22GB | | |
| 560 | Xwin-LM 70B v0.1 | 46±1 | local | ~28GB | | |
| 561 |  Eurus-2-7B-PRIME | 46±2 | local | ~6.4GB | | |
| 562 | Qwen2 7B Instruct | 46±1 | local | ~5.4GB | +4 | |
| 563 | Command R7B (12-2024) | 46±1 | local | ~6.2GB | | |
| 564 | Granite 20B Code Instruct (8K) | 45±1 | local | ~14GB | | |
| 565 |  Dolphin 2.6 Mixtral 8x7B | 45±1 | local | ~29GB | | |
| 566 | Gemma 3 4B | 45±1 | api $0.063/M +local | ~3.8GB | ~74 tok/s | +11 |
| 567 | Granite 3.0 8B Instruct | 45±1 | local | ~6.6GB | | |
| 568 |  LLaVA-OneVision Qwen2 72B (ov-sft) | 45±3 | local | ~155GB | | |
| 569 | Granite 8B Code Instruct (4K) | 45±1 | local | ~6.6GB | | |
| 570 | Falcon3 7B Instruct | 45±2 | local | ~5.9GB | | |
| 571 | Phi-4-mini 3.8B | 45±1 | api $0.13/M +local | ~4GB | +4 | |
| 572 | Ministral 8B (2410) | 45±1 | local | ~6.2GB | +6 | |
| 573 | Code Llama 70B Python | 45±1 | local | ~28GB | | |
| 574 | Code Tulu 2 34B | 45±2† | local | ~72GB | | |
| 575 | Falcon3 10B Instruct | 45±2 | local | ~7.6GB | ~13 tok/s | |
| 576 | Granite 3.2 8B Instruct | 45±1 | local | ~6.6GB | | |
| 577 | OpenThinker 7B | 45±2† | local | ~6GB | | |
| 578 | Yi 34B Chat | 45±1 | local | ~23GB | +1 | |
| 579 | Hermes 2 Theta Llama 3 8B | 45±2 | local | ~6.6GB | | |
| 580 | Llama 3.1 8B (base) | 45±3 | local | ~6.2GB | | |
| 581 | Gemma 3n E2B IT | 45±2 | local | ~4.1GB | | |
| 582 | Code Llama 70B (base) | 45±1 | local | ~28GB | | |
| 583 | Phi-3.5-mini-instruct | 45±1 | local | ~5.7GB | +4 | |
| 584 | Qwen2.5-VL 3B Instruct | 45±1 | local | ~2.8GB | | |
| 585 | Qwen3-VL 2B Instruct | 45±1 | local | ~2.7GB | | |
| 586 | LFM2.5 1.2B Instruct | 45±2 | local | ~1.3GB | | |
| 587 | Qwen3 1.7B | 45±1 | local | ~2.6GB | ~96 tok/s | +11 |
| 590 | Granite 3.1 8B Instruct | 44±1 | local | ~6.7GB | | |
| 593 | Llama 3.2 11B Vision Instruct | 44±2 | local | ~8.9GB | | |
| 594 | Qwen2.5 3B Instruct | 44±1 | local | ~2.6GB | ~61 tok/s | +5 |
| 595 | LFM2 2.6B | 44±1 | local | ~2GB | | |
| 596 | CogVLM2 Llama3 Chat 19B | 44±3 | local | ~43GB | | |
| 597 | Tulu 2 DPO 70B | 44±1 | local | ~32GB | +1 | |
| 598 | Phi-3-small-8k-instruct | 44±2 | local | ~17GB | | |
| 599 | MiniCPM-V 2.6 | 44±2 | local | ~5.8GB | | |
| 600 | CodeGemma 7B IT (v1.0) | 44±1 | local | ~7.1GB | | |
| 601 | Yi-1.5 9B (base) | 44±2 | local | ~6.8GB | | |
| 603 | Qwen2-Audio 7B Instruct | 44±3 | local | ~6.5GB | | |
| 604 | DeepSeek-Coder 1.3B Instruct | 44±1 | local | ~2.9GB | | |
| 605 | Mathstral 7B v0.1 | 44±1 | local | ~5.6GB | | |
| 606 | Aya 23 8B | 44±3 | local | ~6.2GB | | |
| 607 | Qwen2.5 Coder 1.5B Instruct | 44±1 | local | ~1.8GB | | |
| 608 | WizardLM 70B V1.0 | 44±1 | local | ~32GB | | |
| 609 | SmolLM3 3B | 44±1 | local | ~2.9GB | | |
| 610 | Hermes 3 Llama 3.1 8B | 44±1 | local | ~6.6GB | | |
| 611 | Llama 3 8B Instruct | 44±0 | local | ~6.6GB | +19 | |
| 612 | Command R (03-2024) | 44±1 | local | ~33GB | +6 | |
| 613 | Aya Vision 32B | 44±3 | local | ~71GB | | |
| 614 | WizardCoder Python 13B V1.0 | 43±2 | local | ~15GB | | |
| 615 | OpenChat 3.5 | 43±1 | local | ~6.1GB | | |
| 616 | Granite 34B Code Base (8K) | 43±1 | local | ~24GB | | |
| 617 | speechless-starcoder2-7b | 43±1 | local | ~6.1GB | | |
| 618 | OLMo 2 7B Instruct (1124) | 43±1 | local | ~5.8GB | | |
| 619 | Falcon2 11B (base) | 43±3 | local | ~8.2GB | | |
| 620 |  Snowflake Arctic Instruct | 43±2 | local | ~1007GB | | |
| 621 | Phi-3-vision-128k-instruct | 43±2 | local | ~10GB | | |
| 622 | WizardCoder-15B-V1.0 | 43±1 | local | ~12GB | | |
| 625 | DeepSeek LLM 67B Chat | 43±1 | local | ~31GB | | |
| 626 | Qwen1.5 14B Chat | 43±1 | local | ~11GB | +3 | |
| 627 | Llama 2 70B (base) | 43±1 | local | ~32GB | | |
| 628 | Mixtral 8x7B v0.1 (base) | 43±1 | local | ~29GB | | |
| 629 | Code Llama 34B Python | 43±1 | local | ~23GB | | |
| 630 | DeepSeek-Coder 33B Base | 43±1 | local | ~22GB | | |
| 631 | speechless-coding-7b-16k-tora | 43±1 | local | ~16GB | | |
| 632 | InternVL2-4B | 43±2 | local | ~10GB | | |
| 633 | InternLM2 Chat 20B | 43±1 | local | ~14GB | | |
| 634 | CodeQwen1.5 7B (base) | 42±1 | local | ~6.4GB | | |
| 635 | Nous Hermes 2 Mistral 7B DPO | 42±3 | local | ~6.1GB | | |
| 636 | Mixtral 8x7B Instruct v0.1 | 42±0 | local | ~28GB | +15 | |
| 638 | Granite 4.0 1B | 42±1 | local | ~2.4GB | | |
| 639 | OpenHermes-2.5-Code-290k-13B | 42±1 | local | ~9.8GB | | |
| 640 | Granite 4.0 H 1B | 42±1 | local | ~2.3GB | | |
| 641 | Nous Hermes 2 Mixtral 8x7B DPO | 42±2 | local | ~30GB | | |
| 642 | Molmo 7B-D (0924) | 42±1 | local | ~35GB | | |
| 643 |  Deita 7B v1.0 | 42±2 | local | ~32GB | | |
| 645 | North Micro Vision Instruct (2.4B) | 42±2† | local | ~3.8GB | | |
| 646 | OpenHermes 2.5 Mistral 7B | 42±1 | local | ~5.9GB | +2 | |
| 647 | Code Llama 34B Instruct | 42±1 | local | ~22GB | +1 | |
| 648 | Code Llama 34B (base) | 42±1 | local | ~23GB | +1 | |
| 649 | Starling-LM 7B beta | 42±1 | local | ~6GB | +3 | |
| 650 |  Vicuna 33B v1.3 | 41±1 | local | ~23GB | +2 | |
| 651 | Phi-3-mini-128k-instruct (Apr 2024) | 41±2 | local | ~9.5GB | +1 | |
| 652 | Code Tulu 2 13B | 41±2† | local | ~29GB | | |
| 653 | Bielik 1.5B v3.0 Instruct | 41±3 | local | ~2.4GB | | |
| 654 |  Starling-LM 7B alpha | 41±1 | local | ~6.1GB | +2 | |
| 655 | Granite 3.3 2B Instruct | 41±1† | local | ~3GB | | |
| 656 |  Llama3-OpenBioLLM-8B | 41±2 | local | ~6.6GB | | |
| 657 | Mistral 7B Instruct v0.3 | 41±1 | local | ~5.6GB | +4 | |
| 658 |  LLaVA-v1.6-34B | 41±2 | local | ~24GB | | |
| 659 | LFM2 1.2B | 41±1 | local | ~1.3GB | | |
| 660 | DeepSeek-Coder 6.7B Base | 41±1 | local | ~8.8GB | | |
| 661 | Granite 3.0 2B Instruct | 41±1 | local | ~3.2GB | | |
| 662 | Llama 2 70B Chat | 41±1 | local | ~32GB | +13 | |
| 663 | Llama 3.2 3B Instruct | 41±1 | api $0.12/M +local | ~3.3GB | ~139 tok/s | +11 |
| 664 | Granite 8B Code Base (4K) | 41±1 | local | ~6.6GB | | |
| 665 | StarCoder2 15B | 41±1 | local | ~11GB | | |
| 666 | Llama-3-Instruct-8B-SimPO-v0.2 | 40±1 | local | ~6.6GB | | |
| 667 | Falcon-H1 3B Instruct | 40±3 | local | ~3.4GB | | |
| 668 | Phi-2 (2.7B) | 40±1 | local | ~4.7GB | +2 | |
| 669 | speechless-code-mistral-7b-v1.0 | 40±1 | local | ~17GB | | |
| 670 | Qwen-14B-Chat | 40±1 | local | ~11GB | | |
| 671 | Code Llama 13B Instruct | 40±1 | local | ~15GB | | |
| 672 | Tulu 2 13B | 40±2 | local | ~9.8GB | | |
| 673 | Falcon-180B-Chat | 40±2 | local | ~115GB | | |
| 674 |  MistralHermes-CodePro-7B-v1 | 40±1 | local | ~6.1GB | | |
| 676 | Phi-3-mini-4k-instruct (Apr 2024) | 40±2 | local | ~6GB | | |
| 677 | CodeGemma 7B (v1.0) | 40±1 | local | ~7.1GB | | |
| 678 | Code Llama 13B Python | 40±1 | local | ~9.8GB | | |
| 679 |  Apertus 8B Instruct (2509) | 40±3 | local | ~6.3GB | | |
| 680 | DeepSeek-V2 Lite Chat (16B-A2.4B) | 40±1 | local | ~11GB | | |
| 681 | Granite 3B Code Instruct (2K) | 40±1 | local | ~3.7GB | | |
| 682 | Tulu 2 DPO 13B | 39±1 | local | ~9.8GB | | |
| 684 |  BioMistral-7B | 39±3 | local | ~6.1GB | | |
| 685 | SOLAR 10.7B Instruct v1.0 | 39±1 | local | ~8.3GB | +2 | |
| 686 | Gemma 2 2B IT | 39±1 | local | ~3.2GB | +4 | |
| 687 | Code Tulu 2 7B | 39±2† | local | ~16GB | | |
| 688 | Flan-T5 XXL | 39±1 | local | ~49GB |  +12 | |
| 689 | Yi-1.5 6B (base) | 39±3 | local | ~5.1GB | | |
| 690 | Dolphin 2.2.1 Mistral 7B | 39±2 | local | ~5.9GB | | |
| 691 | Gemma 1.1 7B IT | 39±1 | local | ~7.1GB | +2 | |
| 692 | Granite 3.1 2B Instruct | 39±1 | local | ~3.1GB | | |
| 693 | Phi-3.5-vision-instruct | 39±2 | local | ~10GB | | |
| 694 | Granite 20B Code Base (8K) | 39±1 | local | ~15GB | | |
| 695 | Code Llama 7B Python | 39±1 | local | ~5.8GB | | |
| 696 | Granite 3.2 2B Instruct | 39±1 | local | ~3GB | | |
| 697 | Qwen1.5 7B Chat | 39±1 | local | ~9.7GB | +4 | |
| 698 | Zephyr 7B alpha | 38±1 | local | ~6.1GB | | |
| 699 | Code Llama 13B (base) | 38±1 | local | ~9.2GB | +1 | |
| 700 | Falcon3 3B Instruct | 38±3 | local | ~3.4GB | | |
| 701 | Qwen2.5 Coder 0.5B Instruct | 38±1 | local | ~1GB | | |
| 702 | Vicuna 13B v1.5 | 38±1 | local | ~9.8GB | | |
| 703 | Yi 1.5 6B Chat | 38±2 | local | ~4.4GB | | |
| 704 | WizardLM 13B V1.2 | 38±1 | local | ~9.8GB | | |
| 705 | Qwen2-VL 2B Instruct | 38±1 | local | ~2.1GB | | |
| 706 | Qwen2.5 1.5B Instruct | 38±1 | local | ~1.8GB | ~98 tok/s | +6 |
| 707 | Tulu 2 7B | 38±2 | local | ~5.8GB | | |
| 709 | Mistral 7B Instruct v0.2 | 38±1 | local | ~5.8GB | +9 | |
| 710 | Code Llama 7B Instruct | 38±1 | local | ~5.5GB | | |
| 711 | LFM2 700M | 38±1 | local | ~1GB | | |
| 712 | Qwen3.5 0.8B | 38±1 | local | ~1.4GB | ~94 tok/s | +2 |
| 714 | EXAONE 3.5 2.4B Instruct | 38±2 | local | ~3GB | | |
| 715 |  xDAN-L1-Chat-RL-v1 | 37±1 | local | ~6.1GB | | |
| 717 | Mistral 7B v0.1 (base) | 37±1 | local | ~5.8GB | +4 | |
| 718 | Llama 3 8B (base) | 37±1 | local | ~6.6GB | | |
| 719 | Flan-T5 XL | 37±1 | local | ~14GB | +8 | |
| 720 | Gemma 7B (base) | 37±1 | local | ~7.1GB | +3 | |
| 721 | Tulu 2 DPO 7B | 37±1 | local | ~5.8GB | | |
| 722 | Vicuna 13B v1.3 | 37±2 | local | ~29GB | | |
| 723 | Granite 3.1 3B-A800M Instruct | 37±2† | local | ~3.6GB | ~23 tok/s | |
| 724 | Granite 3.0 3B-A800M Instruct | 37±1 | local | ~3.7GB | | |
| 725 | DeepSeek R1 Distill Qwen 1.5B | 37±2 | local | ~1.9GB | +1 | |
| 726 | StarCoder2 7B | 37±1 | local | ~5.4GB | | |
| 728 | DeepSeek LLM 7B Chat | 37±2† | local | ~5.7GB | | |
| 729 | Falcon 40B (base) | 36±1 | local | ~27GB | +2 | |
| 730 | StarCoder 15.5B | 36±1 | local | ~11GB | | |
| 731 | Nous-Hermes-13b | 36±2 | local | ~9.8GB | | |
| 732 | Eurus-2-7B-SFT | 36±3† | local | ~6.4GB | | |
| 733 | Qwen3 0.6B | 36±1 | local | ~1.9GB | ~182 tok/s | +5 |
| 735 | Code Llama 7B (base) | 36±1 | local | ~5.5GB | +2 | |
| 736 | Llama 2 13B (base) | 36±1 | local | ~9.3GB | +4 | |
| 737 | Llama 2 13B Chat | 36±1 | local | ~15GB | +10 | |
| 738 | Zephyr 7B beta | 35±1 | local | ~6.1GB | +3 | |
| 739 |  Guanaco 33B | 35±1 | local | ~70GB | | |
| 740 | StarCoder2 3B | 35±1 | local | ~3.4GB | | |
| 741 | Granite 3B Code Base (2K) | 35±1 | local | ~3.7GB | | |
| 742 | StarCoderBase 15.5B | 35±1 | local | ~12GB | | |
| 743 | Vicuna 7B v1.5 | 35±1 | local | ~5.8GB | +1 | |
| 744 | ChatGLM3 6B 128K | 35±4 | local | ~5.8GB | | |
| 746 | OLMo 2 1B Instruct (0425) | 35±1 | local | ~2.4GB | | |
| 747 | Vicuna 7B v1.3 | 35±2 | local | ~5.8GB | | |
| 748 |  Stable Code 3B | 35±1 | local | ~4.8GB | | |
| 749 | Orca 2 13B | 35±3 | local | ~9.8GB | | |
| 751 | DeepSeek-Coder 1.3B Base | 34±1 | local | ~2.4GB | | |
| 752 | Qwen1.5 4B Chat | 34±2 | local | ~4.1GB | | |
| 753 | Mistral 7B Instruct v0.1 | 34±1 | local | ~5.9GB | +3 | |
| 754 | Gemma 7B IT | 34±1 | local | ~7.1GB | +5 | |
| 757 | CodeGemma 2B (v1.0) | 34±1 | local | ~3.2GB | | |
| 758 | Qwen2 1.5B Instruct | 34±1 | local | ~1.8GB | +1 | |
| 759 | Falcon-H1 1.5B Instruct | 33±4 | local | ~2.4GB | | |
| 760 |  Ghost 7B Alpha | 33±2 | local | ~6.1GB | | |
| 761 | Gemma 3 1B | 33±1 | local | ~1.7GB | ~98 tok/s | +7 |
| 762 | SmolLM2 1.7B Instruct | 33±1 | local | ~3GB | | |
| 763 | LFM2 350M | 33±1 | local | ~1.7GB | | |
| 764 | Qwen-7B (base) | 33±1 | local | ~6.6GB | | |
| 767 | StarCoderBase 7B | 33±1 | local | ~6GB | | |
| 769 | Llama 3.2 1B Instruct | 33±1 | api $0.020/M +local | ~1.6GB | ~28 tok/s | +13 |
| 770 | Phi-1.5 (1.3B) | 33±1 | local | ~2.4GB | | |
| 772 | Qwen-7B-Chat | 32±1 | local | ~6.6GB | | |
| 773 | Llama 2 7B Chat | 32±1 | local | ~8.8GB | +13 | |
| 775 | OLMo 7B Instruct | 32±2 | local | ~16GB | | |
| 776 | Llama 2 7B (base) | 32±1 | local | ~8.5GB | +6 | |
| 778 | Flan-T5 Large | 32±1 | local | ~4.8GB | +5 | |
| 779 | Granite 3.1 1B-A400M Instruct | 31±2† | local | ~2.3GB | ~20 tok/s | |
| 780 | Falcon 40B Instruct | 31±2 | local | ~27GB | | |
| 781 | Gemma 2B (base) | 31±1 | local | ~3.2GB | +3 | |
| 782 | Gemma 1.1 2B IT | 31±1 | local | ~3.2GB | +1 | |
| 783 | Falcon 7B (base) | 31±1 | local | ~5.9GB | +4 | |
| 785 |  Baichuan2-7B-Base | 30±1 | local | ~17GB | | |
| 787 | NexusRaven V2 13B | 30±3 | local | ~9.3GB | | |
| 789 | ChatGLM3-6B | 30±2 | local | ~5.8GB | | |
| 790 | Qwen2.5 0.5B Instruct | 30±2 | local | ~1GB | +1 | |
| 791 | OLMo 7B | 30±1† | local | ~30GB | | |
| 792 |  OpenLLaMA 7B v2 | 30±1 | local | ~9.1GB | | |
| 793 | StableLM-3B-4E1T | 30±1 | local | ~3.3GB | | |
| 794 | StarCoderBase 3B | 30±1 | local | ~3.4GB | | |
| 796 | StableLM-Base-Alpha 7B v2 | 30±1 | local | ~30GB | | |
| 797 | Granite 3.0 1B-A400M Instruct | 30±1 | local | ~2.3GB | | |
| 798 | OPT-66B | 29±1 | local | ~141GB | +2 | |
| 799 |  GPT4All-13B-Snoozy | 29±2 | local | ~56GB | | |
| 801 | EuroLLM 1.7B Instruct | 29±3 | local | ~2.4GB | | |
| 802 | Granite 4.0 H 350M | 29±1† | local | ~1.7GB | | |
| 804 | Gemma 2B IT | 29±1 | local | ~3.1GB | +3 | |
| 805 |  BLOOM 176B | 29±1 | local | ~372GB | +1 | |
| 806 | Granite 4.0 350M | 28±1 | local | ~1.7GB | | |
| 807 | OPT-30B | 28±1 | local | ~64GB | +5 | |
| 809 |  GPT-NeoX-20B | 28±1 | local | ~14GB | +1 | |
| 810 | Qwen2 0.5B Instruct | 28±2 | local | ~1GB | | |
| 811 |  RedPajama-INCITE 7B Base | 28±1 | local | ~16GB | | |
| 813 | ChatGLM2-6B | 28±2 | local | ~15GB | | |
| 816 | Falcon3 1B Instruct | 26±4 | local | ~2.4GB | | |
| 817 | Qwen1.5 1.8B Chat | 26±2 | local | ~2.7GB | | |
| 819 | OPT-13B (13 billion parameters) | 26±1 | local | ~29GB | +4 | |
| 820 | Falcon 7B Instruct | 26±2 | local | ~5.9GB | +1 | |
| 822 |  OpenAssistant SFT-4 Pythia 12B | 25±1 | local | ~9.5GB | +1 | |
| 823 | Flan-T5 Base | 25±1 | local | ~2.5GB | +3 | |
| 824 | RedPajama-INCITE Base 3B v1 | 25±3 | local | ~3.4GB | | |
| 826 | Pythia 12B | 25±1 | local | ~9.5GB | +1 | |
| 827 | OpenLLaMA 3B v2 | 25±1 | local | ~4.2GB | | |
| 828 | GPT-J 6B | 25±1 | local | ~14GB | +6 | |
| 829 | ChatGLM-6B | 25±1 | local | ~16GB | +3 | |
| 830 | StableLM-Base-Alpha 3B v2 | 24±1 | local | ~13GB | | |
| 831 | OPT-6.7B | 24±1 | local | ~16GB | +5 | |
| 834 | Qwen1.5 0.5B Chat | 23±3 | local | ~1.7GB | | |
| 835 | Pythia 6.9B | 23±1 | local | ~6.1GB | +2 | |
| 837 | Phi-1 (1.3B) | 22±4 | local | ~2.4GB | | |
| 839 | SmolLM2 360M Instruct | 22±3 | local | ~1GB | | |
| 840 | FastChat-T5 3B | 21±2 | local | ~8.5GB | | |
| 841 | OLMo 1B | 21±2† | local | ~2.2GB | | |
| 842 | BLOOM 7.1B | 21±1 | local | ~7.1GB | +7 | |
| 843 | Pythia 2.8B | 20±1 | local | ~3.4GB | | |
| 844 | Pythia 2.8B (deduped) | 20±1 | local | ~3.4GB | | |
| 845 | Falcon-RW 1B | 20±2 | local | ~4.3GB | | |
| 846 | OPT-2.7B | 19±1 | local | ~7.1GB | +3 | |
| 847 | StableLM-Tuned-Alpha 7B | 18±2 | local | ~6.9GB | | |
| 848 | GPT-Neo 2.7B | 18±1 | local | ~13GB | +8 | |
| 850 | T5 Large | 17±2 | local | ~4.6GB | | |
| 851 | Gemma 3 270M IT | 17±4 | local | ~1.1GB | | |
| 852 | Flan-T5 Small | 16±3 | local | ~1.8GB | | |
| 853 | BLOOM 3B | 16±1 | local | ~4GB | +2 | |
| 856 | OPT-1.3B | 14±1 | local | ~4.3GB | +5 | |
| 857 | T5 Base | 14±3 | local | ~2.4GB | | |
| 858 | Pythia 1.4B | 14±1 | local | ~2.4GB | | |
| 859 | Pythia 1B | 12±1 | local | ~2.2GB | | |
| 862 | GPT-Neo 1.3B | 12±2 | local | ~7.1GB | +3 | |
| 863 | T5 Small | 12±3 | local | ~1.8GB | | |
| 864 | BLOOM 1.7B | 11±1 | local | ~3.1GB | +1 | |
| 865 |  Zyte-1B | 11±2 | local | ~2.2GB | | |
| 867 | StableLM-Base-Alpha 7B | 8±1 | local | ~35GB | | |
| 868 | Pythia 410M | 8±2 | local | ~1.8GB | | |
| 870 | BLOOM 1.1B | 7±1 | local | ~2.6GB | | |
| 871 | GPT-2 XL (1.5B) | 7±1 | local | ~2.8GB | ~75 tok/s | +9 |
| 874 | OPT-350M | 1±1 | local | ~2.2GB | +1 | |
| 876 | StableLM-Base-Alpha 3B | 1±2 | local | ~17GB | | |
| 877 | BLOOM 560M | 0±1 | local | ~2.1GB | +1 | |
| 878 | Pythia 160M | -2±2 | local | ~1.6GB | | |
| 879 | GPT-2 Large (774M) | -3±1 | local | ~2GB | ~137 tok/s | +3 |
| 880 | OPT-125M | -3±1 | local | ~1.8GB | +1 | |
| 881 | GPT-Neo 125M | -3±2 | local | ~2.1GB | | |
| 882 | GPT-2 Medium (355M) | -5±1 | local | ~1.8GB | +3 | |
| 883 | Pythia 70M | -13±2 | local | ~1.6GB | | |
| 884 | GPT-2 Small (124M) | -13±2 | local | ~1.6GB | +3 | |
| — | Grok Build 0.1 | no reviewed results | api $1.25/M | |||
| — |  K2 Horizon 7B | not scored yet | local | ~8GB | ||
| — | LFM2.5 24B-A2B | not yet calibrated | local | ~17GB | | |
| — | LFM3 1.7B | not yet calibrated | local | ~2.6GB | |
+ 203 more don't fit your machine or budget · show them
+ 227 models with fewer than 3 results left off, since one or two tests say little about overall ability · show them
919 of 1227 catalogue models listed here · sources and licences
* new or thinly tested: ranked on too few benchmark results · † only the maker’s own numbers so far · ~RAM is estimated; full context may need more memory