Sync 2026-10-16: 156 models

README.md

@@ -36,7 +36,7 @@ table at **[modelfit.io/data](https://modelfit.io/data/)**; the canonical
 machine-readable source is
 **[modelfit.io/api/dataset](https://modelfit.io/api/dataset/)**.
 
-- **156 models** across 27 families (116 with a registry-verified local build, 40 cloud-only APIs tracked for comparison)
+- **156 models** across 27 families (116 with a registry-verified local build, 23 open-weight models without a consumer-tier Ollama build, 17 closed cloud APIs tracked for comparison)
 - **License:** CC BY 4.0: reuse freely, including commercially, with attribution
 - **Updated:** see the `updated` field in `models.json` (regenerated from the live endpoint)
 

generate.mjs

@@ -1,7 +1,7 @@
 // Regenerate models.json + models.csv from the live ModelFit dataset endpoint.
 // Usage: node generate.mjs   (then commit the changes)
 // The canonical source is https://modelfit.io/api/dataset/ (CC BY 4.0).
-import { writeFileSync } from 'node:fs';
+import { readFileSync, writeFileSync } from 'node:fs';
 
 const SRC = 'https://modelfit.io/api/dataset/';
 const UA = 'modelfit-hardware-dataset/1.0 (+https://modelfit.io/data/)';
@@ -26,4 +26,13 @@ const csv = [cols.join(',')]
   .join('\n');
 writeFileSync('models.csv', csv + '\n');
 
-console.log(`Wrote ${data.models.length} models (updated ${data.updated}) to models.json + models.csv`);
+// Keep the headline counts of the Hugging Face card in step with the data.
+const total = data.models.length;
+const local = data.models.filter((r) => r.runsLocally).length;
+const closed = data.models.filter((r) => !r.runsLocally && !r.openWeights).length;
+const families = new Set(data.models.map((r) => r.family)).size;
+const line = `- **${total} models** across ${families} families (${local} with a registry-verified local build, ${total - local - closed} open-weight models without a consumer-tier Ollama build, ${closed} closed cloud APIs tracked for comparison)`;
+const card = readFileSync('hf-card.md', 'utf8');
+writeFileSync('hf-card.md', card.replace(/^- \*\*\d+ models\*\* across .*$/m, line));
+
+console.log(`Wrote ${total} models (updated ${data.updated}) to models.json + models.csv + hf-card.md`);

models.csv

@@ -8,17 +8,17 @@ Llama 3.1 8B Instruct,Llama,8,Q4_K_M,12,6.5,,true,true,false,ollama|llama.cpp|lm
 Llama 3.1 8B Instruct (Q8),Llama,8,Q8_0,16,8,,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding",ollama run llama3.1:8b-instruct-q8_0
 Llama 3.1 8B Instruct (Q5),Llama,8,Q5_K_M,12,8,,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding",ollama run llama3.1:8b-instruct-q5_K_M
 Llama 3.1 70B Instruct,Llama,70,Q4_K_M,64,42,,true,true,false,ollama|llama.cpp|lm-studio,"Quality, Coding",ollama run llama3.1:70b-instruct-q4_K_M
-Mistral 7B Instruct,Mistral,7,Q4_K_M,8,5.5,,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding",ollama run mistral:7b-instruct-q4_K_M
-Mistral 7B Instruct (Q5),Mistral,7,Q5_K_M,8,4.8,,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding",ollama run mistral:7b-instruct-q5_K_M
-Mistral 7B Instruct (Q8),Mistral,7,Q8_0,16,7.2,,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding",ollama run mistral:7b-instruct-q8_0
+Mistral 7B Instruct,Mistral,7,Q4_K_M,8,5.5,,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding",ollama run mistral:7b-instruct-v0.3-q4_K_M
+Mistral 7B Instruct (Q5),Mistral,7,Q5_K_M,8,4.8,,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding",ollama run mistral:7b-instruct-v0.3-q5_K_M
+Mistral 7B Instruct (Q8),Mistral,7,Q8_0,16,7.2,,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding",ollama run mistral:7b-instruct-v0.3-q8_0
 Mixtral 8x7B Instruct,Mistral,46.7,Q4_K_M,48,30,,true,true,false,ollama|llama.cpp|lm-studio,"Coding, Quality",ollama run mixtral:8x7b
 Mistral Nemo 12B,Mistral,12,Q4_K_M,16,9.5,,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Translation",ollama run mistral-nemo:12b
 Mistral Nemo 12B (Q8),Mistral,12,Q8_0,24,12.1,,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Translation",ollama run mistral-nemo:12b-instruct-2407-q8_0
 Gemma 2 2B Instruct,Gemma,2,Q4_K_M,4,1.8,,true,true,false,ollama|llama.cpp|lm-studio,Chat,ollama run gemma2:2b-instruct-q4_K_M
 Gemma 2 9B Instruct,Gemma,9,Q4_K_M,12,7,,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding",ollama run gemma2:9b-instruct-q4_K_M
 Gemma 2 27B Instruct,Gemma,27,Q4_K_M,32,21,,true,true,false,ollama|llama.cpp|lm-studio,"Quality, Coding",ollama run gemma2:27b-instruct-q4_K_M
-Phi-3 Mini 3.8B,Phi,3.8,Q4_K_M,6,3.2,,true,true,false,ollama|llama.cpp|lm-studio,"Coding, Chat",ollama run phi3:mini
-Phi-3 Medium 14B,Phi,14,Q4_K_M,16,11,,true,true,false,ollama|llama.cpp|lm-studio,"Coding, Quality",ollama run phi3:medium
+Phi-3 Mini 3.8B,Phi,3.8,Q4_K_M,6,3.2,,true,true,false,ollama|llama.cpp|lm-studio,"Coding, Chat",ollama run phi3:3.8b-mini-4k-instruct-q4_K_M
+Phi-3 Medium 14B,Phi,14,Q4_K_M,16,11,,true,true,false,ollama|llama.cpp|lm-studio,"Coding, Quality",ollama run phi3:14b-medium-4k-instruct-q4_K_M
 Phi-4 14B,Phi,14,Q4_K_M,24,11.5,,true,true,false,ollama|llama.cpp|lm-studio,"Coding, Quality",ollama run phi4:14b-q4_K_M
 Phi-4 14B (Q8),Phi,14,Q8_0,24,14.5,,true,true,false,ollama|llama.cpp|lm-studio,"Coding, Quality",ollama run phi4:14b-q8_0
 Qwen2.5 Coder 7B,Qwen,7,Q4_K_M,8,5.5,,true,true,false,ollama|llama.cpp|lm-studio,Coding,ollama run qwen2.5-coder:7b
@@ -42,7 +42,7 @@ DeepSeek-R1 671B,DeepSeek,671,Q4_K_M,512,380,,true,true,false,ollama|llama.cpp|l
 Qwen3 8B,Qwen,8,Q4_K_M,12,6.5,,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding",ollama run qwen3:8b-q4_K_M
 Qwen3 8B (Q8),Qwen,8,Q8_0,16,8.1,,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding",ollama run qwen3:8b-q8_0
 Qwen3 14B,Qwen,14,Q4_K_M,16,11,,true,true,false,ollama|llama.cpp|lm-studio,"Coding, Quality",ollama run qwen3:14b-q4_K_M
-Qwen3 30B,Qwen,30,Q4_K_M,32,22,,true,true,false,ollama|llama.cpp|lm-studio,"Quality, Coding",ollama run qwen3:30b
+Qwen3 30B,Qwen,30,Q4_K_M,32,22,,true,true,false,ollama|llama.cpp|lm-studio,"Quality, Coding",ollama run qwen3:30b-a3b-q4_K_M
 Qwen3 30B (Q8),Qwen,30,Q8_0,48,30.3,,true,true,false,ollama|llama.cpp|lm-studio,"Quality, Coding",ollama run qwen3:30b-a3b-q8_0
 Gemma 3 4B Instruct,Gemma,4,Q4_K_M,6,3.5,,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding",ollama run gemma3:4b
 Gemma 3 4B Instruct (Q8),Gemma,4,Q8_0,8,3.9,,true,true,false,ollama|llama.cpp|lm-studio,"Chat, Coding",ollama run gemma3:4b-it-q8_0

models.json

@@ -10,7 +10,7 @@
     "note": "A model fits a device when its estimatedLoadGb is within the memory budget: ~70% of unified memory up to 32GB, scaling linearly to ~85% at 128GB and above (high-RAM Macs can wire more memory to the GPU). Q4_K_M needs roughly 0.6 GB per billion parameters.",
     "fieldSemantics": "runsLocally = a registry-verified Ollama build fits at least one consumer RAM tier tracked here (up to 256GB). openWeights = the weights are publicly downloadable. Some open-weight giants (e.g. NVIDIA Nemotron 3 Ultra, 550B, ~190GB at 2-bit) are openWeights: true but runsLocally: false because no consumer tier fits them."
   },
-  "updated": "2026-10-03",
+  "updated": "2026-10-13",
   "counts": {
     "models": 156,
     "local": 116,
@@ -206,7 +206,7 @@
         "lm-studio"
       ],
       "bestFor": "Chat, Coding",
-      "ollamaCommand": "ollama run mistral:7b-instruct-q4_K_M"
+      "ollamaCommand": "ollama run mistral:7b-instruct-v0.3-q4_K_M"
     },
     {
       "model": "Mistral 7B Instruct (Q5)",
@@ -225,7 +225,7 @@
         "lm-studio"
       ],
       "bestFor": "Chat, Coding",
-      "ollamaCommand": "ollama run mistral:7b-instruct-q5_K_M"
+      "ollamaCommand": "ollama run mistral:7b-instruct-v0.3-q5_K_M"
     },
     {
       "model": "Mistral 7B Instruct (Q8)",
@@ -244,7 +244,7 @@
         "lm-studio"
       ],
       "bestFor": "Chat, Coding",
-      "ollamaCommand": "ollama run mistral:7b-instruct-q8_0"
+      "ollamaCommand": "ollama run mistral:7b-instruct-v0.3-q8_0"
     },
     {
       "model": "Mixtral 8x7B Instruct",
@@ -377,7 +377,7 @@
         "lm-studio"
       ],
       "bestFor": "Coding, Chat",
-      "ollamaCommand": "ollama run phi3:mini"
+      "ollamaCommand": "ollama run phi3:3.8b-mini-4k-instruct-q4_K_M"
     },
     {
       "model": "Phi-3 Medium 14B",
@@ -396,7 +396,7 @@
         "lm-studio"
       ],
       "bestFor": "Coding, Quality",
-      "ollamaCommand": "ollama run phi3:medium"
+      "ollamaCommand": "ollama run phi3:14b-medium-4k-instruct-q4_K_M"
     },
     {
       "model": "Phi-4 14B",
@@ -832,7 +832,7 @@
         "lm-studio"
       ],
       "bestFor": "Quality, Coding",
-      "ollamaCommand": "ollama run qwen3:30b"
+      "ollamaCommand": "ollama run qwen3:30b-a3b-q4_K_M"
     },
     {
       "model": "Qwen3 30B (Q8)",