AA-LCR v1.1
Extracting, reasoning about and synthesizing information in long documents.
Long-context reasoningReported independently by Artificial Analysis386 models run by Artificial AnalysisUpdated 8 October 2026
How it is run
AA evaluates documents ranging from 10k to 100k tokens using the cl100k_base tokenizer and reports its v1.1 score. Highest published effort per release; the table retains its snapshot date.
What it cannot tell you
Effort settings and evaluation versions matter. These source-specific results are kept separate from Vals AI and vendor reports. A missing score stays blank.
- Leader
- Kimi K388.67% · Artificial Analysis results, grouped here by release
- Top-ten spread
- 5.00%Leader minus tenth place, or last place in smaller fields.
- Median
- 50.84%of the 386 rows published
- No published result
- 4of 46 models on this site
Artificial Analysis results by release
Highest published effort per model release, grouped here from AA’s public results. Grouping ranks are derived here; source settings and model-publisher reports stay separate.
| # | Model | Score | Effort |
|---|---|---|---|
| 1 | Kimi K3 | 88.67% | max |
| 2 | Step 5 Preview | 88.33% | Not specified |
| 3 | MiMo-V2.6-Pro | 86.33% | Not specified |
| 4 | Claude Fable 5.1 | 85.33% | max |
| 5 | Claude Opus 5.5 | 84.67% | max |
| 6 | GPT-5.5 | 84.33% | xhigh |
| 7 | GPT-5.6 Sol | 84.00% | max |
| 7 | DeepSeek V4.1 Flash | 84.00% | max |
| 9 | GPT-6 Sol | 83.67% | max |
| 9 | GPT-5.6 Luna | 83.67% | max |
| 11 | GPT-6 Luna | 83.33% | max |
| 11 | Muse Glimmer | 83.33% | high |
| 11 | GPT-5.3 Codex | 83.33% | xhigh |
| 11 | Solar Mini 4 | 83.33% | Not specified |
| 15 | GPT-6.1 Sol | 83.00% | max |
| 15 | GPT-5.6 Terra | 83.00% | max |
| 15 | MiniMax-M3 | 83.00% | Not specified |
| 15 | Muse Spark 1.3 | 83.00% | max |
| 19 | Claude Sonnet 5.5 | 82.67% | max |
| 19 | GPT-5.2 | 82.67% | xhigh |
| 19 | Claude Haiku 5.5 | 82.67% | max |
| 22 | GPT-5.2 Codex | 82.33% | xhigh |
| 22 | Claude Fable 5 | 82.33% | max |
| 24 | Claude Sonnet 5 | 82.00% | max |
| 24 | Qwen3.8 27B | 82.00% | xhigh |
| 24 | GPT-5.4 | 82.00% | xhigh |
| 24 | Gemini 3.1 Pro Preview | 82.00% | Not specified |
| 28 | Nex-N2-Pro | 81.67% | Not specified |
| 28 | Gemini 3.7 Flash | 81.67% | high |
| 30 | Gemini 3.8 Flash | 81.33% | high |
| 30 | DeepSeek V4 Flash Vision | 81.33% | max |
| 32 | Kimi K2.6 | 81.00% | Not specified |
| 32 | Grok 4.6 | 81.00% | xhigh |
| 34 | GPT-6 Astra | 80.67% | max |
| 34 | Qwen3.6 Max Preview | 80.67% | Not specified |
| 36 | DeepSeek V4 Pro 0813 | 80.33% | max |
| 36 | Qwen3.8 2.4T A95B | 80.33% | Not specified |
| 36 | Qwen3.8 Max (0902) | 80.33% | Not specified |
| 39 | GLM 5.3 Flash | 80.00% | max |
| 39 | Claude Sonnet 4.6 | 80.00% | max |
| 39 | GPT-5.1 | 80.00% | high |
| 39 | Gemini 3.6 Flash | 80.00% | high |
| 39 | K2 Horizon 375B A23B | 80.00% | Not specified |
| 44 | MiMo-V2.5-Pro | 79.67% | Not specified |
| 44 | Qwen3.8-Flash-Next | 79.67% | Not specified |
| 44 | GLM-5.3 | 79.67% | max |
| 44 | DeepSeek V4 Flash 0731 | 79.67% | max |
| 44 | Gemini 4 Argon | 79.67% | high |
| 49 | Apodex 1.1 | 79.33% | Not specified |
| 49 | Claude Opus 5 | 79.33% | max |
| 49 | Nemotron 3 Ultra 550B A55B | 79.33% | Not specified |
| 49 | Grok 4.5 | 79.33% | high |
| 49 | Kimi K2.7 Code | 79.33% | Not specified |
| 54 | Muse Spark 1.2 | 79.00% | xhigh |
| 54 | Qwen3.7 Max | 79.00% | Not specified |
| 54 | Hy3 | 79.00% | Not specified |
| 57 | Claude Opus 4.7 | 78.67% | max |
| 58 | Ling-3.0-flash-VL | 78.33% | Not specified |
| 58 | JT-4.1 Flash 236B A21B | 78.33% | Not specified |
| 58 | MiniMax-M2.7 | 78.33% | Not specified |
| 58 | Qwen3.8 Max | 78.33% | Not specified |
| 58 | GLM-5.2 | 78.33% | max |
| 58 | Qwen3.6 Plus | 78.33% | Not specified |
| 64 | GPT-5 | 78.20% | high |
| 65 | Claude Opus 4.6 | 78.00% | max |
| 65 | Kimi K2.5 | 78.00% | Not specified |
| 65 | Muse Spark | 78.00% | Not specified |
| 68 | Qwen3.5 27B | 77.67% | Not specified |
| 68 | Claude Opus 4.8 | 77.67% | max |
| 68 | Muse Spark 1.1 | 77.67% | xhigh |
| 71 | Inkling | 77.33% | xhigh |
| 71 | Qwen3.5 397B A17B | 77.33% | Not specified |
| 71 | Claude Opus 4.5 | 77.33% | Not specified |
| 74 | GPT-5.4 mini | 77.00% | xhigh |
| 75 | Grok 4.7 | 76.67% | xhigh |
| 75 | GPT-5.4 nano | 76.67% | xhigh |
| 77 | Quasar 438B | 76.33% | max |
| 78 | Gemini 3.5 Flash-Lite | 76.00% | high |
| 78 | Motif 3 | 76.00% | Not specified |
| 78 | Gemini 3 Pro Preview | 76.00% | high |
| 78 | Claude 4.1 Opus | 76.00% | Not specified |
| 82 | GLM-5 | 75.67% | Not specified |
| 82 | Inkling Small | 75.67% | Not specified |
| 84 | MiMo-V2-Omni | 75.00% | Not specified |
| 85 | Grok Build 0.1 0616 | 74.67% | Not specified |
| 85 | o3 | 74.67% | Not specified |
| 85 | DeepSeek V4 Pro 0424 | 74.67% | max |
| 85 | Motif 3 (Beta) | 74.67% | Not specified |
| 89 | MiMo-V2.6-Flash | 74.33% | Not specified |
| 89 | Gemini 3.1 Flash-Lite | 74.33% | high |
| 89 | Qwen3 Max Thinking | 74.33% | Not specified |
| 89 | DeepSeek V4 Flash 0420 | 74.33% | max |
| 89 | Claude 4.5 Haiku | 74.33% | Not specified |
| 94 | Solar Pro 4 | 74.00% | Not specified |
| 94 | Grok 4.1 Fast | 74.00% | Not specified |
| 96 | GLM-5.1 | 73.67% | Not specified |
| 96 | Step 3.7 Flash | 73.67% | high |
| 96 | Ling-3.0-flash-Fin | 73.67% | Not specified |
| 99 | Gemini 3.5 Flash | 73.33% | high |
| 99 | MiniMax-M2.5 | 73.33% | Not specified |
| 99 | MiMo-V2-Omni-0327 | 73.33% | Not specified |
| 102 | MiMo-V2.5 | 73.00% | Not specified |
| 102 | Grok 4.3 | 73.00% | high |
| 102 | Qwen3.7 Plus | 73.00% | Not specified |
| 102 | Ling 3.0 Flash | 73.00% | Not specified |
| 102 | KAT Coder Pro V2 | 73.00% | Not specified |
| 107 | GPT-5 mini | 72.33% | high |
| 107 | Claude 4.5 Sonnet | 72.33% | Not specified |
| 109 | Qwen3.5 35B A3B | 72.00% | Not specified |
| 109 | Kimi K2 Thinking | 72.00% | Not specified |
| 109 | Qwen3 235B A22B 2507 (Reasoning) | 72.00% | Not specified |
| 112 | GLM-5-Turbo | 71.67% | Not specified |
| 112 | Doubao Seed Code | 71.67% | Not specified |
| 112 | K2 Horizon MoVA 36B A4B | 71.67% | Not specified |
| 112 | GPT-5 Codex | 71.67% | high |
| 112 | Mercury 2.5 | 71.67% | Not specified |
| 112 | Solar Open2 250B | 71.67% | Not specified |
| 118 | MiMo-V2-Flash (Feb 2026) | 71.33% | Not specified |
| 119 | GLM-4.7 | 71.00% | Not specified |
| 120 | GLM 5V Turbo | 70.33% | Not specified |
| 120 | Claude 4 Sonnet | 70.33% | Not specified |
| 122 | DeepSeek V3.2 Speciale | 70.00% | Not specified |
| 122 | A.X-K2 | 70.00% | Not specified |
| 122 | GPT-5.5 Instant (June 2026) | 70.00% | Not specified |
| 122 | Ring-2.6-1T | 70.00% | xhigh |
| 126 | K2 Horizon 7B | 69.67% | Not specified |
| 126 | Gemma 4 31B | 69.67% | Not specified |
| 128 | GPT-5.1 Codex | 69.33% | high |
| 128 | DeepSeek V3.1 Terminus | 69.33% | Not specified |
| 128 | Claude 4 Opus | 69.33% | Not specified |
| 128 | Mistral Medium 3.5 | 69.33% | high |
| 132 | Gemini 2.5 Pro | 69.00% | Not specified |
| 133 | GPT-4.1 | 68.33% | Not specified |
| 133 | MiMo-V2-Pro | 68.33% | Not specified |
| 135 | Grok 4 | 68.00% | Not specified |
| 136 | Grok 4.20 0309 | 67.67% | Not specified |
| 136 | MiniMax-M2.1 | 67.67% | Not specified |
| 138 | Qwen3.6 27B | 66.67% | Not specified |
| 138 | GPT-5.1 Codex mini | 66.67% | high |
| 140 | GPT-5.5 Instant (May 2026) | 66.33% | Not specified |
| 141 | JT-35B-Flash | 65.67% | Not specified |
| 141 | Nemotron 3 Super 120B A12B | 65.67% | Not specified |
| 141 | Gemma 4 26B A4B | 65.67% | Not specified |
| 144 | Gemini 2.5 Flash | 65.33% | Not specified |
| 144 | G9v3-39A5B | 65.33% | Not specified |
| 146 | o1 | 65.00% | Not specified |
| 146 | LongCat 2.0 | 65.00% | Not specified |
| 146 | GPT-5 (ChatGPT) | 65.00% | Not specified |
| 149 | Hy3-preview | 64.67% | Not specified |
| 150 | Qwen3.6 35B A3B | 64.33% | Not specified |
| 150 | MiniMax-M2 | 64.33% | Not specified |
| 152 | Nova 2.0 Pro Preview | 64.00% | medium |
| 153 | Qwen3.5 Omni Plus | 63.67% | Not specified |
| 153 | Qwen3 VL 235B A22B (Reasoning) | 63.67% | Not specified |
| 153 | Qwen3 Next 80B A3B (Reasoning) | 63.67% | Not specified |
| 156 | Step 3.5 Flash 2603 | 63.00% | Not specified |
| 157 | K2 Horizon 3.7B | 62.33% | Not specified |
| 158 | Qwen3 Max Thinking (Preview) | 62.00% | Not specified |
| 159 | Qwen3.5 122B A10B | 61.33% | Not specified |
| 159 | Seed-OSS-36B-Instruct | 61.33% | Not specified |
| 159 | Qwen3 30B A3B 2507 (Reasoning) | 61.33% | Not specified |
| 159 | K-EXAONE | 61.33% | Not specified |
| 163 | o4-mini | 61.00% | high |
| 164 | Nemotron 3.5 Lightning | 60.33% | Not specified |
| 164 | Nova 2.0 Lite | 60.33% | high |
| 164 | Ling 3.0 Tiny | 60.33% | Not specified |
| 167 | K-EXAONE 2.0 0803 | 60.00% | Not specified |
| 167 | Gemini 2.5 Flash Preview (Sep '25) | 60.00% | Not specified |
| 169 | Nova 2.0 Omni | 59.67% | medium |
| 170 | MiniCPM5-2B | 59.33% | Not specified |
| 171 | Grok 3 | 58.00% | Not specified |
| 172 | DeepSeek R1 (Jan '25) | 57.67% | Not specified |
| 172 | MiniMax M1 80k | 57.67% | Not specified |
| 174 | K2 Think V2 | 57.00% | high |
| 175 | DeepSeek V3.1 | 56.67% | Not specified |
| 176 | GPT-4o (ChatGPT) | 56.00% | Not specified |
| 177 | DeepSeek R1 0528 (May '25) | 55.67% | Not specified |
| 178 | Gemini 3 Flash Preview | 55.33% | Not specified |
| 178 | Qwen3 VL 32B (Reasoning) | 55.33% | Not specified |
| 178 | Grok 3 mini Reasoning | 55.33% | high |
| 181 | EXAONE 4.5 33B | 54.67% | Not specified |
| 181 | Ring-1T | 54.67% | Not specified |
| 183 | GLM-4.6 | 54.00% | Not specified |
| 184 | Kimi K2 | 53.00% | Not specified |
| 184 | Kimi K2 0905 | 53.00% | Not specified |
| 184 | Magistral Medium 1.2 | 53.00% | Not specified |
| 184 | Grok Code Fast 1 | 53.00% | Not specified |
| 188 | GLM-4.5 | 52.67% | Not specified |
| 188 | Qwen3 Next 80B A3B Instruct | 52.67% | Not specified |
| 188 | Command A+ | 52.67% | Not specified |
| 191 | Qwen3.5 Omni Flash | 52.00% | Not specified |
| 191 | gpt-oss-120b | 52.00% | high |
| 193 | Claude 3.7 Sonnet | 51.67% | Not specified |
| 194 | Qwen3 Max | 50.00% | Not specified |
| 194 | Llama 4 Maverick | 50.00% | Not specified |
| 194 | Step 3.5 Flash | 50.00% | Not specified |
| 197 | Gemini 2.5 Flash-Lite Preview (Sep '25) | 49.67% | Not specified |
| 198 | GPT-4o (Nov '24) | 49.33% | Not specified |
| 199 | Granite 4.2 30B | 49.00% | Not specified |
| 200 | Qwen3 Coder Next | 47.00% | Not specified |
| 201 | GLM-4.5-Air | 46.67% | Not specified |
| 202 | Qwen3.5 9B | 46.00% | Not specified |
| 203 | Qwen3 Coder 480B A35B Instruct | 45.67% | Not specified |
| 203 | DeepSeek V3.2 | 45.67% | Not specified |
| 205 | GPT-5 nano | 45.00% | high |
| 205 | Granite 4.2 8B | 45.00% | Not specified |
| 207 | DeepSeek V3.2 Exp | 44.67% | Not specified |
| 208 | GPT-4.1 mini | 44.00% | Not specified |
| 209 | Mercury 2 | 43.67% | high |
| 210 | o3-mini | 43.00% | high |
| 210 | Qwen3 Max (Preview) | 43.00% | Not specified |
| 212 | Mistral Medium 3.1 | 42.67% | Not specified |
| 213 | GLM-4.7-Flash | 41.67% | Not specified |
| 213 | Ling-2.6-1T | 41.67% | Not specified |
| 215 | G9v3-3B | 41.33% | Not specified |
| 216 | Solar Open 100B | 41.00% | high |
| 216 | NVIDIA Nemotron Nano 12B v2 VL | 41.00% | Not specified |
| 216 | GPT-4o (Aug '24) | 41.00% | Not specified |
| 219 | DeepSeek V3 0324 | 40.67% | Not specified |
| 220 | INTELLECT-3 | 39.67% | Not specified |
| 220 | Nemotron 3 Nano Omni 30B A3B Reasoning | 39.67% | Not specified |
| 222 | Nemotron Cascade 2 30B A3B | 39.33% | Not specified |
| 223 | KAT-Coder-Pro V1 | 38.67% | Not specified |
| 224 | Celeris-1 | 38.00% | Not specified |
| 224 | Trinity Large Thinking | 38.00% | Not specified |
| 226 | North Mini Code | 37.33% | Not specified |
| 226 | Qwen3 4B 2507 (Reasoning) | 37.33% | Not specified |
| 228 | Ling-1T | 36.67% | Not specified |
| 229 | HyperNova 60B 2605 | 36.33% | high |
| 229 | MiMo-V2-Flash | 36.33% | Not specified |
| 231 | Mistral Large 3 | 36.00% | Not specified |
| 232 | Gemma 4 12B | 35.00% | Not specified |
| 232 | K2-V2 | 35.00% | high |
| 234 | gpt-oss-20b | 34.67% | high |
| 234 | Qwen3.5 4B | 34.67% | Not specified |
| 236 | Nova Premier | 34.33% | Not specified |
| 237 | Qwen3 235B A22B 2507 Instruct | 33.90% | Not specified |
| 238 | Qwen3 VL 8B (Reasoning) | 33.33% | Not specified |
| 239 | Qwen3 VL 235B A22B Instruct | 32.67% | Not specified |
| 239 | Qwen3 Coder 30B A3B Instruct | 32.67% | Not specified |
| 241 | Devstral 2 | 32.33% | Not specified |
| 241 | Solar Pro 3 | 32.33% | high |
| 243 | Gemini 2.5 Flash-Lite | 32.00% | Not specified |
| 243 | Devstral Medium | 32.00% | Not specified |
| 243 | Devstral Small (May '25) | 32.00% | Not specified |
| 246 | Mistral Medium 3 | 31.33% | Not specified |
| 246 | Ling 2.6 Flash | 31.33% | Not specified |
| 246 | Gemini 2.0 Flash (Feb '25) | 31.33% | Not specified |
| 249 | DeepSeek V3 (Dec '24) | 29.33% | Not specified |
| 250 | Mistral Small 4 | 28.33% | Not specified |
| 250 | LongCat Flash Lite | 28.33% | Not specified |
| 252 | Kimi Linear 48B A3B Instruct | 28.00% | Not specified |
| 252 | Devstral Small 2 | 28.00% | Not specified |
| 254 | Llama 4 Scout | 27.70% | Not specified |
| 255 | Claude 3 Haiku | 27.67% | Not specified |
| 256 | Claude 3.5 Haiku | 27.33% | Not specified |
| 257 | QwQ 32B | 26.67% | Not specified |
| 258 | Ministral 3 14B | 26.33% | Not specified |
| 258 | Qwen3 30B A3B 2507 Instruct | 26.33% | Not specified |
| 260 | Ministral 3 8B | 25.67% | Not specified |
| 261 | Llama 3.1 Instruct 405B | 25.33% | Not specified |
| 262 | Llama Nemotron Super 49B v1.5 | 24.67% | Not specified |
| 263 | Granite 4.2 3B | 24.33% | Not specified |
| 264 | Grok 4 Fast | 24.00% | Not specified |
| 264 | NVIDIA Nemotron Nano 9B V2 | 24.00% | Not specified |
| 264 | Gemma 4 E4B | 24.00% | Not specified |
| 267 | Granite 4.1 30B | 23.67% | Not specified |
| 268 | Grok 4.20 0309 v2 | 23.33% | Not specified |
| 269 | Cogito v2.1 | 22.67% | Not specified |
| 270 | Mistral Small 3.1 | 22.33% | Not specified |
| 270 | Hermes 4 - Llama-3.1 405B | 22.33% | Not specified |
| 272 | Tri-21B-think Preview | 21.67% | Not specified |
| 272 | NVIDIA Nemotron 3 Nano 4B | 21.67% | Not specified |
| 274 | Command A | 21.33% | Not specified |
| 274 | Apriel-v1.5-15B-Thinker | 21.33% | Not specified |
| 274 | Qwen3 VL 4B (Reasoning) | 21.33% | Not specified |
| 277 | Ring-flash-2.0 | 21.00% | Not specified |
| 277 | Nova Pro | 21.00% | Not specified |
| 279 | Mistral Small 3.2 | 20.33% | Not specified |
| 279 | GPT-4.1 nano | 20.33% | Not specified |
| 281 | DiffusionGemma 26B A4B | 19.67% | Not specified |
| 282 | Magistral Small 1.2 | 19.33% | Not specified |
| 282 | Jamba 1.7 Large | 19.33% | Not specified |
| 284 | Nova Lite | 19.00% | Not specified |
| 285 | Devstral Small (Jul '25) | 18.67% | Not specified |
| 286 | Llama 3.3 Nemotron Super 49B v1 | 18.33% | Not specified |
| 287 | Llama 3.1 Instruct 8B | 18.00% | Not specified |
| 288 | Ling-flash-2.0 | 17.67% | Not specified |
| 289 | GLM-4.6V | 17.00% | Not specified |
| 289 | Ministral 3 3B | 17.00% | Not specified |
| 291 | Qwen3 VL 8B Instruct | 16.67% | Not specified |
| 292 | Gemma 4 E2B | 16.33% | Not specified |
| 293 | Jamba 1.7 Mini | 16.00% | Not specified |
| 294 | Tri-21B-Think | 15.67% | Not specified |
| 294 | Llama 3.3 Instruct 70B | 15.67% | Not specified |
| 296 | Phi-4 Mini Instruct | 15.33% | Not specified |
| 297 | DeepSeek R1 0528 Qwen3 8B | 15.00% | Not specified |
| 298 | Motif-2-12.7B-Reasoning | 14.67% | Not specified |
| 298 | Mi:dm K 2.5 Pro Preview | 14.67% | Not specified |
| 300 | Qwen3.5 2B | 14.00% | Not specified |
| 300 | Qwen3 VL 4B Instruct | 14.00% | Not specified |
| 302 | HyperCLOVA X SEED Think (32B) | 13.67% | Not specified |
| 303 | Granite 4.1 8B | 13.33% | Not specified |
| 304 | Nova Micro | 13.00% | Not specified |
| 304 | JT-MINI | 13.00% | Not specified |
| 306 | Llama 3.2 Instruct 11B (Vision) | 12.67% | Not specified |
| 307 | Granite 4.0 H Small | 11.33% | Not specified |
| 307 | Qwen3 4B 2507 Instruct | 11.33% | Not specified |
| 309 | Falcon-H1R-7B | 11.00% | Not specified |
| 310 | NVIDIA Nemotron 3 Nano 30B A3B | 10.67% | Not specified |
| 311 | DeepSeek R1 Distill Qwen 14B | 10.33% | Not specified |
| 312 | DeepSeek R1 Distill Llama 70B | 10.00% | Not specified |
| 313 | Mi:dm K 2.5 Pro | 9.00% | Not specified |
| 314 | DeepSeek R1 Distill Qwen 32B | 8.67% | Not specified |
| 315 | Gemma 3 12B Instruct | 8.33% | Not specified |
| 315 | Llama 3.1 Nemotron Instruct 70B | 8.33% | Not specified |
| 317 | Qwen3.5 0.8B | 8.00% | Not specified |
| 318 | Gemma 3 27B Instruct | 7.33% | Not specified |
| 319 | Granite 4.0 Micro | 7.00% | Not specified |
| 319 | MiniCPM-V 4.6 1.3B | 7.00% | Not specified |
| 319 | Granite 4.0 H 1B | 7.00% | Not specified |
| 322 | Gemma 3 4B Instruct | 6.67% | Not specified |
| 322 | Llama 3.2 Instruct 1B | 6.67% | Not specified |
| 324 | K2 Horizon 0.9B | 6.33% | Not specified |
| 324 | Jamba Reasoning 3B | 6.33% | Not specified |
| 326 | MiniCPM5-1B | 6.00% | Not specified |
| 326 | Granite 4.1 3B | 6.00% | Not specified |
| 326 | Granite 4.0 1B | 6.00% | Not specified |
| 329 | LFM2.5-2.6B | 5.67% | Not specified |
| 330 | Hermes 4 - Llama-3.1 70B | 4.67% | Not specified |
| 331 | Llama 3.2 Instruct 3B | 4.33% | Not specified |
| 332 | Mistral Large 2 (Jul '24) | 2.00% | Not specified |
| 333 | DeepSeek R1 Distill Qwen 1.5B | 1.67% | Not specified |
| 334 | GLM-4.5V | 0.00% | Not specified |
| 334 | Gemma 3 270M | 0.00% | Not specified |
| 334 | Qwen3 Omni 30B A3B Instruct | 0.00% | Not specified |
| 334 | DeepSeek R1 Distill Llama 8B | 0.00% | Not specified |
| 334 | Qwen3 0.6B | 0.00% | Not specified |
| 334 | LFM2 1.2B | 0.00% | Not specified |
| 334 | Magistral Small 1 | 0.00% | Not specified |
| 334 | Qwen3 4B | 0.00% | Not specified |
| 334 | Mistral 7B Instruct | 0.00% | Not specified |
| 334 | Qwen3 1.7B | 0.00% | Not specified |
| 334 | Olmo 3 7B Think | 0.00% | Not specified |
| 334 | Sarvam M | 0.00% | high |
| 334 | Olmo 3.1 32B Think | 0.00% | Not specified |
| 334 | LFM2.5-VL-1.6B | 0.00% | Not specified |
| 334 | Qwen3 30B A3B | 0.00% | Not specified |
| 334 | Molmo2-8B | 0.00% | Not specified |
| 334 | Qwen3 235B A22B | 0.00% | Not specified |
| 334 | Solar Pro 2 | 0.00% | high |
| 334 | Qwen3 14B | 0.00% | Not specified |
| 334 | Sarvam 105B | 0.00% | high |
| 334 | LFM2.5-1.2B-Instruct | 0.00% | Not specified |
| 334 | Apertus 70B Instruct | 0.00% | Not specified |
| 334 | OLMo 2 32B | 0.00% | Not specified |
| 334 | Granite 4.0 350M | 0.00% | Not specified |
| 334 | LFM2.5-1.2B-Thinking | 0.00% | Not specified |
| 334 | Magistral Medium 1 | 0.00% | Not specified |
| 334 | LFM2 2.6B | 0.00% | Not specified |
| 334 | Sarvam 30B | 0.00% | high |
| 334 | Nanbeige4.1-3B | 0.00% | Not specified |
| 334 | Tiny Aya Global | 0.00% | Not specified |
| 334 | Exaone 4.0 1.2B | 0.00% | Not specified |
| 334 | Llama 3 Instruct 8B | 0.00% | Not specified |
| 334 | Apertus 8B Instruct | 0.00% | Not specified |
| 334 | Olmo 3 7B Instruct | 0.00% | Not specified |
| 334 | Olmo 3.1 32B Instruct | 0.00% | Not specified |
| 334 | Qwen3 8B | 0.00% | Not specified |
| 334 | Granite 4.0 H 350M | 0.00% | Not specified |
| 334 | Gemma 3n E4B Instruct | 0.00% | Not specified |
| 334 | Gemma 3n E2B Instruct | 0.00% | Not specified |
| 334 | Llama 3 Instruct 70B | 0.00% | Not specified |
| 334 | Step3 VL 10B | 0.00% | Not specified |
| 334 | Molmo 7B-D | 0.00% | Not specified |
| 334 | Qwen3 32B | 0.00% | Not specified |
| 334 | Llama 3.1 Nemotron Nano 4B v1.1 | 0.00% | Not specified |
| 334 | Qwen3 Omni 30B A3B (Reasoning) | 0.00% | Not specified |
| 334 | Olmo 3 32B Think | 0.00% | Not specified |
| 334 | Gemma 3 1B Instruct | 0.00% | Not specified |
| 334 | Reka Flash 3 | 0.00% | Not specified |
| 334 | LFM2.5-8B-A1B | 0.00% | Not specified |
| 334 | Mistral Small 3 | 0.00% | Not specified |
| 334 | Phi-4 | 0.00% | Not specified |
| 334 | OLMo 2 7B | 0.00% | Not specified |
| 334 | LFM2 24B A2B | 0.00% | Not specified |
No published result tracked
4 of the 46 models on this site have no published number on this benchmark, so their cells in the home table stay blank rather than reading as zero.
Read this with
Other versions of the same evaluation. Versions are not comparable: the task set changes between them.
Sources
Every page used for this benchmark, read on 17 October 2026.
- Benchmark site or the page that publishes itartificialanalysis.ai/evaluations/artificial-analysis-long-context-reasoning
- Independent resultsartificialanalysis.ai/evaluations/artificial-analysis-long-context-reasoning