Specialised model rankings
102 current Index positionsScored language models with no coding or terminal evidence — chat SKUs, translation, roleplay, OCR and other niche products. Coding and terminal models are ranked on the All board. Official publisher releases only; quantizations and repacks stay discoverable, unranked.
102 shown
2Current-model rank | GPT-6 SolNew OpenAI | 88.41 public benchmark observation across 1 board; Index basis Measured; 20% support; score interval 67.3–100.0. | 1M 128K max output Exact context: 1,050,000 tokens; exact maximum output: 128,000 tokens | — No Arena rating | $10.00 |
6Current-model rank | 86.51 public benchmark observation across 1 board; Index basis Measured; 20% support; score interval 65.4–100.0. | 1M 131K max output Exact context: 1,000,000 tokens; exact maximum output: 131,072 tokens | — No Arena rating | $6.00 | |
27 Decreased by 2 Current-model rank | GPT-6 LunaNew OpenAI | 78.43 public benchmark observations across 3 boards; Index basis Measured; 23% support; score interval 57.5–99.3. | 1M 128K max output Exact context: 1,050,000 tokens; exact maximum output: 128,000 tokens | — No Arena rating | $0.50 |
49 Increased by 1 Current-model rank | InclusionAILocal model | 67.81 public benchmark observation across 1 board; Index basis Measured; 20% support; score interval 46.7–88.9. | 262K 33K max output Exact context: 262,144 tokens; exact maximum output: 32,768 tokens | — No Arena rating | $0.18 |
58Current-model rank | InclusionAILocal model | 66.01 public benchmark observation across 1 board; Index basis Measured; 20% support; score interval 44.9–87.1. | 262K 236K max output Exact context: 262,144 tokens; exact maximum output: 235,929 tokens | — No Arena rating | $0.18 |
59Current-model rank | OpenAI | 65.3publisher-corpus-prior over 223 held-out anchors (58%); 1 specialist-board observation at 2.1% support (42%); shrunk 9.1 toward the measured corpus; Index basis Measured; 2% support; score interval 42.5–88.1. | 1M 128K max output Exact context: 1,050,000 tokens; exact maximum output: 128,000 tokens | — No Arena rating | $180 |
66 Decreased by 1 Current-model rank | AisingaporeLocal model | 62.6finetune of qwen/qwen3.6-27b · finetune (100%); Index basis Inherited; 8% support; score interval 45.7–79.5. | — — max output Exact context: 0 tokens | — No Arena rating | Self-hosted: weights only · no listed API price |
67 Decreased by 1 Current-model rank | 62.6finetune of qwen/qwen3.6-27b · finetune (100%); Index basis Inherited; 8% support; score interval 45.7–79.5. | — — max output Exact context: 0 tokens | — No Arena rating | Self-hosted: weights only · no listed API price | |
69 Decreased by 1 Current-model rank | 62.6finetune of qwen/qwen3.6-27b · finetune (100%); Index basis Inherited; 8% support; score interval 45.7–79.5. | — — max output Exact context: 0 tokens | — No Arena rating | Self-hosted: weights only · no listed API price | |
74 Decreased by 10 Current-model rank | Dots Studio | 61.5global-corpus-prior over 223 held-out anchors (58%); 1 specialist-board observation at 2.1% support (42%); shrunk 6 toward the measured corpus; Index basis Measured; 2% support; score interval 38.7–84.3. | 512K 461K max output Exact context: 512,000 tokens; exact maximum output: 460,800 tokens | — No Arena rating | Free |
76Current-model rank | Meituan LongcatLocal model | 60.9global-corpus-prior over 223 held-out anchors (8%); launch card against 4 resolved peers on 12 rows (92%); exceeds every named peer on 2 of 12 rows; optimism haircut 0 from cross-lab-probe; shrunk 0.3 toward the measured corpus; Index basis Estimated; 40% support; score interval 52.3–69.5. | — — max output Exact context: 0 tokens | — No Arena rating | Self-hosted: weights only · no listed API price |
83Current-model rank | MicrosoftLocal model | 58.4finetune of qwen/qwen3.5-27b · finetune (100%); Index basis Inherited; 31% support; score interval 41.5–75.3. | — — max output Exact context: 0 tokens | — No Arena rating | Self-hosted: weights only · no listed API price |
91Current-model rank | Darkc0deLocal model | 57.2finetune of darkc0de/XORTRON.CriminalComputing.2026.27B.NEXT · finetune (100%); Index basis Inherited; 9% support; score interval 38.8–75.6. | — — max output Exact context: 0 tokens | — No Arena rating | Self-hosted: weights only · no listed API price |
92 Decreased by 1 Current-model rank | OpenbmbLocal model | 56.91 public benchmark observation across 1 board; Index basis Measured; 20% support; score interval 35.8–78.0. | — — max output Exact context: 0 tokens | — No Arena rating | Self-hosted: weights only · no listed API price |
93 Increased by 43 Current-model rank | Inception Labs | 56.81 public benchmark observation across 1 board; Index basis Measured; 20% support; score interval 35.7–77.9. | 260K 66K max output Exact context: 260,000 tokens; exact maximum output: 65,536 tokens | — No Arena rating | $0.15 |
96 Decreased by 1 Current-model rank | InclusionAILocal model | 56.0finetune of qwen/qwen3-235b-a22b-thinking-2507 · finetune (100%); Index basis Inherited; 28% support; score interval 39.1–72.9. | — — max output Exact context: 0 tokens | — No Arena rating | Self-hosted: weights only · no listed API price |
97 Decreased by 1 Current-model rank | 56.0finetune of qwen/qwen3-235b-a22b-thinking-2507 · finetune (100%); Index basis Inherited; 28% support; score interval 39.1–72.9. | — — max output Exact context: 0 tokens | — No Arena rating | Self-hosted: weights only · no listed API price | |
99 Decreased by 1 Current-model rank | Mistral AILocal model | 55.81 public benchmark observation across 1 board; Index basis Measured; 20% support; score interval 34.7–76.9. | 262K 210K max output Exact context: 262,144 tokens; exact maximum output: 209,715 tokens | — No Arena rating | $0.60 |
101 Decreased by 1 Current-model rank | MicrosoftLocal model | 55.1finetune of deepseek/deepseek-r1 · finetune (100%); Index basis Inherited; 32% support; score interval 38.2–72.0. | — — max output Exact context: 0 tokens | — No Arena rating | Self-hosted: weights only · no listed API price |
102 Decreased by 1 Current-model rank | 55.0finetune of qwen/qwen3.5-35b-a3b · finetune (100%); Index basis Inherited; 31% support; score interval 38.1–71.9. | — — max output Exact context: 0 tokens | — No Arena rating | Unlisted: no listed API price | |
118 Decreased by 2 Current-model rank | TencentLocal model | 49.2finetune of google/gemma-4-E4B-it · finetune (100%); Index basis Inherited; 8% support; score interval 40.3–58.2. | — — max output Exact context: 0 tokens | — No Arena rating | Unlisted: no listed API price |
121 Decreased by 2 Current-model rank | InclusionAI | 47.9finetune of qwen/qwen3.5-9b · finetune (100%); Index basis Inherited; 15% support; score interval 32.8–63.0. | — — max output Exact context: 0 tokens | — No Arena rating | Unlisted: no listed API price |
122 Decreased by 2 Current-model rank | MicrosoftLocal model | 47.9finetune of qwen/qwen3.5-9b · finetune (100%); Index basis Inherited; 15% support; score interval 32.8–63.0. | — — max output Exact context: 0 tokens | — No Arena rating | Self-hosted: weights only · no listed API price |
126 Decreased by 2 Current-model rank | AisingaporeLocal model | 47.1finetune of google/gemma-4-E2B-it · finetune (100%); Index basis Inherited; 8% support; score interval 38.1–56.0. | — — max output Exact context: 0 tokens | — No Arena rating | Self-hosted: weights only · no listed API price |
132 Decreased by 2 Current-model rank | MicrosoftLocal model | 45.5finetune of Qwen/Qwen3.5-4B · finetune (100%); Index basis Inherited; 15% support; score interval 31.1–59.8. | — — max output Exact context: 0 tokens | — No Arena rating | Self-hosted: weights only · no listed API price |
