<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Model Tested</title><link>https://modeltested.com/</link><description>We test AI models on real work and tell you which ones are worth using.</description><item><title>The best local LLMs right now</title><link>https://modeltested.com/best-local-llms/</link><guid>https://modeltested.com/best-local-llms/</guid><pubDate>Sun, 11 Oct 2026 12:00:00 GMT</pubDate><category>Rankings</category><description>We tested 56 AI models you can run on your own computer. Here's the best one for every graphics card and Mac.</description></item><item><title>Best local LLM for coding: we ran the code</title><link>https://modeltested.com/best-local-llms/coding/</link><guid>https://modeltested.com/best-local-llms/coding/</guid><pubDate>Sun, 11 Oct 2026 12:00:00 GMT</pubDate><category>Rankings</category><description>56 models, 30 programming jobs, each checked against tests the model never saw. Here's who writes code that works.</description></item><item><title>Decision models vs. regular LLMs: 14 free models beat the best one</title><link>https://modeltested.com/best-decision-models/</link><guid>https://modeltested.com/best-decision-models/</guid><pubDate>Sun, 11 Oct 2026 12:00:00 GMT</pubDate><category>Rankings</category><description>68 models, 56 real decisions: routing tickets, spotting phishing, flagging risky code. Accuracy and how honest their confidence is.</description></item><item><title>We made our document test harder. Here's what changed</title><link>https://modeltested.com/posts/document-test-harder/</link><guid>https://modeltested.com/posts/document-test-harder/</guid><pubDate>Sun, 11 Oct 2026 12:00:00 GMT</pubDate><category>News</category><description>Most models used to ace it. Now only a handful do, and the rankings moved a lot.</description></item><item><title>We tested Mistral Large 4 ("Le Chonk"). It's brilliant when it stops thinking</title><link>https://modeltested.com/posts/le-chonk-tested/</link><guid>https://modeltested.com/posts/le-chonk-tested/</guid><pubDate>Sun, 11 Oct 2026 12:00:00 GMT</pubDate><category>News</category><description>Every program it finished was correct, and it beat every local model at reading documents. But on 12 of 30 coding jobs it never gave an answer at all.</description></item><item><title>gpt-oss-120b review: one of the best local models we've tested</title><link>https://modeltested.com/models/gpt-oss-120b/</link><guid>https://modeltested.com/models/gpt-oss-120b/</guid><pubDate>Sun, 11 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 96 out of 100, #2 of 56. It solved 28 of 30 coding jobs and scored 99 on reading documents. Runs on a Mac with 96 GB.</description></item><item><title>Muse Glimmer 30B review: one of the best local models we've tested</title><link>https://modeltested.com/models/muse-glimmer-30b/</link><guid>https://modeltested.com/models/muse-glimmer-30b/</guid><pubDate>Sun, 11 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 94 out of 100, #3 of 56. It solved 27 of 30 coding jobs and scored 98 on reading documents. Runs on a 24 GB graphics card or a Mac with 32 GB.</description></item><item><title>Ling 3.0 Flash review: a solid all-rounder</title><link>https://modeltested.com/models/ling-3-0-flash/</link><guid>https://modeltested.com/models/ling-3-0-flash/</guid><pubDate>Sun, 11 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 94 out of 100, #4 of 56. It solved 28 of 30 coding jobs and scored 95 on reading documents. Runs on a Mac with 128 GB.</description></item><item><title>GLM 4.5V review: a solid all-rounder</title><link>https://modeltested.com/models/glm-4-5v/</link><guid>https://modeltested.com/models/glm-4-5v/</guid><pubDate>Sun, 11 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 93 out of 100, #5 of 56. It solved 26 of 30 coding jobs and scored 99 on reading documents. Runs on a Mac with 96 GB.</description></item><item><title>Ling 3.0 Flash VL review: a solid all-rounder</title><link>https://modeltested.com/models/ling-3-0-flash-vl/</link><guid>https://modeltested.com/models/ling-3-0-flash-vl/</guid><pubDate>Sun, 11 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 92 out of 100, #6 of 56. It solved 26 of 30 coding jobs and scored 98 on reading documents. Runs on a Mac with 128 GB.</description></item><item><title>Laguna XS 2.1 review: a solid all-rounder</title><link>https://modeltested.com/models/laguna-xs-2-1/</link><guid>https://modeltested.com/models/laguna-xs-2-1/</guid><pubDate>Sun, 11 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 82 out of 100, #12 of 56. It solved 22 of 30 coding jobs and scored 91 on reading documents. Runs on a 24 GB graphics card or a Mac with 48 GB.</description></item><item><title>GLM 4.5 Air review: decent, with trade-offs</title><link>https://modeltested.com/models/glm-4-5-air/</link><guid>https://modeltested.com/models/glm-4-5-air/</guid><pubDate>Sun, 11 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 76 out of 100, #18 of 56. It solved 21 of 30 coding jobs and scored 82 on reading documents. Runs on a Mac with 96 GB.</description></item><item><title>GLM 4.6V review: decent, with trade-offs</title><link>https://modeltested.com/models/glm-4-6v/</link><guid>https://modeltested.com/models/glm-4-6v/</guid><pubDate>Sun, 11 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 73 out of 100, #19 of 56. It solved 22 of 30 coding jobs and scored 73 on reading documents. Runs on a Mac with 96 GB.</description></item><item><title>Laguna S 2.1 review: decent, with trade-offs</title><link>https://modeltested.com/models/laguna-s-2-1/</link><guid>https://modeltested.com/models/laguna-s-2-1/</guid><pubDate>Sun, 11 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 64 out of 100, #25 of 56. It solved 21 of 30 coding jobs and scored 59 on reading documents. Runs on a Mac with 128 GB.</description></item><item><title>Nex-N2.5-Mini review: not one we'd recommend right now</title><link>https://modeltested.com/models/nex-n2-5-mini/</link><guid>https://modeltested.com/models/nex-n2-5-mini/</guid><pubDate>Sun, 11 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 57 out of 100, #27 of 56. It solved 8 of 30 coding jobs and scored 88 on reading documents. Runs on a 24 GB graphics card or a Mac with 48 GB.</description></item><item><title>Devstral 2 2512 review: not one we'd recommend right now</title><link>https://modeltested.com/models/devstral-2512/</link><guid>https://modeltested.com/models/devstral-2512/</guid><pubDate>Sun, 11 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 53 out of 100, #28 of 56. It solved 16 of 30 coding jobs and scored 53 on reading documents. Runs on a Mac with 128 GB.</description></item><item><title>Hunyuan A13B Instruct review: not one we'd recommend right now</title><link>https://modeltested.com/models/hunyuan-a13b-instruct/</link><guid>https://modeltested.com/models/hunyuan-a13b-instruct/</guid><pubDate>Sun, 11 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 52 out of 100, #29 of 56. It solved 11 of 30 coding jobs and scored 67 on reading documents. Runs on a Mac with 96 GB.</description></item><item><title>Qwen2.5 72B Instruct review: not one we'd recommend right now</title><link>https://modeltested.com/models/qwen-2-5-72b-instruct/</link><guid>https://modeltested.com/models/qwen-2-5-72b-instruct/</guid><pubDate>Sun, 11 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 46 out of 100, #31 of 56. It solved 12 of 30 coding jobs and scored 52 on reading documents. Runs on a Mac with 64 GB.</description></item><item><title>Llama 3.3 70B Instruct review: not one we'd recommend right now</title><link>https://modeltested.com/models/llama-3-3-70b-instruct/</link><guid>https://modeltested.com/models/llama-3-3-70b-instruct/</guid><pubDate>Sun, 11 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 46 out of 100, #33 of 56. It solved 14 of 30 coding jobs and scored 45 on reading documents. Runs on a Mac with 64 GB.</description></item><item><title>Command A review: not one we'd recommend right now</title><link>https://modeltested.com/models/command-a/</link><guid>https://modeltested.com/models/command-a/</guid><pubDate>Sun, 11 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 45 out of 100, #34 of 56. It solved 10 of 30 coding jobs and scored 56 on reading documents. Runs on a Mac with 96 GB.</description></item><item><title>Llama 3.1 70B Instruct review: not one we'd recommend right now</title><link>https://modeltested.com/models/llama-3-1-70b-instruct/</link><guid>https://modeltested.com/models/llama-3-1-70b-instruct/</guid><pubDate>Sun, 11 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 39 out of 100, #41 of 56. It solved 10 of 30 coding jobs and scored 44 on reading documents. Runs on a Mac with 64 GB.</description></item><item><title>Llama 3.3 8B Instruct (Q4, Mac) review: not one we'd recommend right now</title><link>https://modeltested.com/models/llama-3-3-8b-q4-mac/</link><guid>https://modeltested.com/models/llama-3-3-8b-q4-mac/</guid><pubDate>Sun, 11 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 23 out of 100, #47 of 56. It solved 4 of 30 coding jobs and scored 33 on reading documents. Runs on an 8 GB graphics card or a Mac with 16 GB.</description></item><item><title>Granite 4.0 Micro (Q4, Mac) review: not one we'd recommend right now</title><link>https://modeltested.com/models/granite-4-0-h-micro-mac/</link><guid>https://modeltested.com/models/granite-4-0-h-micro-mac/</guid><pubDate>Sun, 11 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 22 out of 100, #49 of 56. It solved 4 of 30 coding jobs and scored 31 on reading documents. Runs on an 8 GB graphics card or a Mac with 16 GB.</description></item><item><title>Qwen2.5 VL 72B Instruct review: not one we'd recommend right now</title><link>https://modeltested.com/models/qwen2-5-vl-72b-instruct/</link><guid>https://modeltested.com/models/qwen2-5-vl-72b-instruct/</guid><pubDate>Sun, 11 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 14 out of 100, #53 of 56. It solved 4 of 30 coding jobs and scored 14 on reading documents. Runs on a Mac with 64 GB.</description></item><item><title>Reka Edge review: not one we'd recommend right now</title><link>https://modeltested.com/models/reka-edge/</link><guid>https://modeltested.com/models/reka-edge/</guid><pubDate>Sun, 11 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 8 out of 100, #54 of 56. It solved 0 of 30 coding jobs and scored 16 on reading documents. Runs on an 8 GB graphics card or a Mac with 16 GB.</description></item><item><title>Llama 3.2 1B Instruct (Q4, Mac) review: not one we'd recommend right now</title><link>https://modeltested.com/models/llama-3-2-1b-q4-mac/</link><guid>https://modeltested.com/models/llama-3-2-1b-q4-mac/</guid><pubDate>Sun, 11 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 5 out of 100, #56 of 56. It solved 0 of 30 coding jobs and scored 9 on reading documents. Runs on an 8 GB graphics card or a Mac with 16 GB.</description></item><item><title>Why we don't publish speed numbers</title><link>https://modeltested.com/posts/why-no-speed-numbers/</link><guid>https://modeltested.com/posts/why-no-speed-numbers/</guid><pubDate>Sat, 10 Oct 2026 12:00:00 GMT</pubDate><category>News</category><description>How fast a model runs depends on your exact computer and app, so any number we gave would be wrong for most people.</description></item><item><title>Q4 or Q8: which version of a model should you download?</title><link>https://modeltested.com/posts/q4-vs-q8/</link><guid>https://modeltested.com/posts/q4-vs-q8/</guid><pubDate>Sat, 10 Oct 2026 12:00:00 GMT</pubDate><category>Guides</category><description>What the letters mean, how much quality you give up, and how to pick the right one for your computer.</description></item><item><title>How to check how much memory your graphics card or Mac has</title><link>https://modeltested.com/posts/check-your-gpu-memory/</link><guid>https://modeltested.com/posts/check-your-gpu-memory/</guid><pubDate>Sat, 10 Oct 2026 12:00:00 GMT</pubDate><category>Guides</category><description>The one number that decides which local models you can run, and where to find it on Windows, Mac and Linux.</description></item><item><title>How to run a local LLM with Ollama</title><link>https://modeltested.com/posts/run-a-local-llm-with-ollama/</link><guid>https://modeltested.com/posts/run-a-local-llm-with-ollama/</guid><pubDate>Sat, 10 Oct 2026 12:00:00 GMT</pubDate><category>Guides</category><description>Install one app, type one command, and you're chatting with an AI model on your own computer. No account needed.</description></item><item><title>Qwen3.6 27B review: one of the best local models we've tested</title><link>https://modeltested.com/models/qwen3-6-27b/</link><guid>https://modeltested.com/models/qwen3-6-27b/</guid><pubDate>Sat, 10 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 97 out of 100, #1 of 56. It solved 29 of 30 coding jobs and scored 97 on reading documents. Runs on a 24 GB graphics card or a Mac with 32 GB.</description></item><item><title>gpt-oss-20b review: a solid all-rounder</title><link>https://modeltested.com/models/gpt-oss-20b/</link><guid>https://modeltested.com/models/gpt-oss-20b/</guid><pubDate>Sat, 10 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 88 out of 100, #7 of 56. It solved 27 of 30 coding jobs and scored 87 on reading documents. Runs on a 16 GB graphics card or a Mac with 24 GB.</description></item><item><title>Qwen3.5-27B review: a solid all-rounder</title><link>https://modeltested.com/models/qwen3-5-27b/</link><guid>https://modeltested.com/models/qwen3-5-27b/</guid><pubDate>Sat, 10 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 88 out of 100, #8 of 56. It solved 23 of 30 coding jobs and scored 100 on reading documents. Runs on a 24 GB graphics card or a Mac with 32 GB.</description></item><item><title>Qwen3.8 27B review: a solid all-rounder</title><link>https://modeltested.com/models/qwen3-8-27b/</link><guid>https://modeltested.com/models/qwen3-8-27b/</guid><pubDate>Sat, 10 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 88 out of 100, #9 of 56. It solved 23 of 30 coding jobs and scored 99 on reading documents. Runs on a 24 GB graphics card or a Mac with 32 GB.</description></item><item><title>Qwen3.5-35B-A3B review: a solid all-rounder</title><link>https://modeltested.com/models/qwen3-5-35b-a3b/</link><guid>https://modeltested.com/models/qwen3-5-35b-a3b/</guid><pubDate>Sat, 10 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 88 out of 100, #10 of 56. It solved 24 of 30 coding jobs and scored 95 on reading documents. Runs on a 24 GB graphics card or a Mac with 48 GB.</description></item><item><title>Qwen3.6 35B A3B review: a solid all-rounder</title><link>https://modeltested.com/models/qwen3-6-35b-a3b/</link><guid>https://modeltested.com/models/qwen3-6-35b-a3b/</guid><pubDate>Sat, 10 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 86 out of 100, #11 of 56. It solved 24 of 30 coding jobs and scored 93 on reading documents. Runs on a 24 GB graphics card or a Mac with 48 GB.</description></item><item><title>Nemotron 3 Nano 30B A3B review: a solid all-rounder</title><link>https://modeltested.com/models/nemotron-3-nano-30b-a3b/</link><guid>https://modeltested.com/models/nemotron-3-nano-30b-a3b/</guid><pubDate>Sat, 10 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 80 out of 100, #13 of 56. It solved 21 of 30 coding jobs and scored 91 on reading documents. Runs on a 24 GB graphics card or a Mac with 32 GB.</description></item><item><title>Gemma 4 31B review: great at code, weaker with documents</title><link>https://modeltested.com/models/gemma-4-31b-it/</link><guid>https://modeltested.com/models/gemma-4-31b-it/</guid><pubDate>Sat, 10 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 79 out of 100, #14 of 56. It solved 28 of 30 coding jobs and scored 65 on reading documents. Runs on a 32 GB graphics card or a Mac with 48 GB.</description></item><item><title>Granite 4.2 8B review: decent, with trade-offs</title><link>https://modeltested.com/models/granite-4-2-8b/</link><guid>https://modeltested.com/models/granite-4-2-8b/</guid><pubDate>Sat, 10 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 77 out of 100, #15 of 56. It solved 23 of 30 coding jobs and scored 78 on reading documents. Runs on an 8 GB graphics card or a Mac with 16 GB.</description></item><item><title>Qwen3 30B A3B review: decent, with trade-offs</title><link>https://modeltested.com/models/qwen3-30b-a3b/</link><guid>https://modeltested.com/models/qwen3-30b-a3b/</guid><pubDate>Sat, 10 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 77 out of 100, #16 of 56. It solved 20 of 30 coding jobs and scored 87 on reading documents. Runs on a 24 GB graphics card or a Mac with 32 GB.</description></item><item><title>Qwen3 VL 30B A3B Thinking review: reads documents well, weaker at code</title><link>https://modeltested.com/models/qwen3-vl-30b-a3b-thinking/</link><guid>https://modeltested.com/models/qwen3-vl-30b-a3b-thinking/</guid><pubDate>Sat, 10 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 77 out of 100, #17 of 56. It solved 18 of 30 coding jobs and scored 93 on reading documents. Runs on a 24 GB graphics card or a Mac with 32 GB.</description></item><item><title>Qwen3 14B review: decent, with trade-offs</title><link>https://modeltested.com/models/qwen3-14b/</link><guid>https://modeltested.com/models/qwen3-14b/</guid><pubDate>Sat, 10 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 72 out of 100, #20 of 56. It solved 19 of 30 coding jobs and scored 81 on reading documents. Runs on a 12 GB graphics card or a Mac with 24 GB.</description></item><item><title>GLM 4.7 Flash review: decent, with trade-offs</title><link>https://modeltested.com/models/glm-4-7-flash/</link><guid>https://modeltested.com/models/glm-4-7-flash/</guid><pubDate>Sat, 10 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 72 out of 100, #21 of 56. It solved 22 of 30 coding jobs and scored 70 on reading documents. Runs on a 24 GB graphics card or a Mac with 48 GB.</description></item><item><title>Qwen3.5-9B review: reads documents well, weaker at code</title><link>https://modeltested.com/models/qwen3-5-9b/</link><guid>https://modeltested.com/models/qwen3-5-9b/</guid><pubDate>Sat, 10 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 72 out of 100, #22 of 56. It solved 14 of 30 coding jobs and scored 97 on reading documents. Runs on an 8 GB graphics card or a Mac with 16 GB.</description></item><item><title>Gemma 4 26B A4B  review: decent, with trade-offs</title><link>https://modeltested.com/models/gemma-4-26b-a4b-it/</link><guid>https://modeltested.com/models/gemma-4-26b-a4b-it/</guid><pubDate>Sat, 10 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 70 out of 100, #23 of 56. It solved 25 of 30 coding jobs and scored 58 on reading documents. Runs on a 24 GB graphics card or a Mac with 32 GB.</description></item><item><title>Qwen3 32B review: decent, with trade-offs</title><link>https://modeltested.com/models/qwen3-32b/</link><guid>https://modeltested.com/models/qwen3-32b/</guid><pubDate>Sat, 10 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 64 out of 100, #24 of 56. It solved 15 of 30 coding jobs and scored 79 on reading documents. Runs on a 24 GB graphics card or a Mac with 48 GB.</description></item><item><title>Nemotron 3.5 Lightning review: not one we'd recommend right now</title><link>https://modeltested.com/models/nemotron-3-5-lightning/</link><guid>https://modeltested.com/models/nemotron-3-5-lightning/</guid><pubDate>Sat, 10 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 57 out of 100, #26 of 56. It solved 13 of 30 coding jobs and scored 72 on reading documents. Runs on a 24 GB graphics card or a Mac with 32 GB.</description></item><item><title>Qwen3 Coder 30B A3B Instruct review: not one we'd recommend right now</title><link>https://modeltested.com/models/qwen3-coder-30b-a3b-instruct/</link><guid>https://modeltested.com/models/qwen3-coder-30b-a3b-instruct/</guid><pubDate>Sat, 10 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 51 out of 100, #30 of 56. It solved 17 of 30 coding jobs and scored 46 on reading documents. Runs on a 24 GB graphics card or a Mac with 32 GB.</description></item><item><title>Qwen3 VL 30B A3B Instruct review: not one we'd recommend right now</title><link>https://modeltested.com/models/qwen3-vl-30b-a3b-instruct/</link><guid>https://modeltested.com/models/qwen3-vl-30b-a3b-instruct/</guid><pubDate>Sat, 10 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 46 out of 100, #32 of 56. It solved 13 of 30 coding jobs and scored 49 on reading documents. Runs on a 24 GB graphics card or a Mac with 32 GB.</description></item><item><title>Ministral 3 14B 2512 review: not one we'd recommend right now</title><link>https://modeltested.com/models/ministral-14b-2512/</link><guid>https://modeltested.com/models/ministral-14b-2512/</guid><pubDate>Sat, 10 Oct 2026 12:00:00 GMT</pubDate><category>Reviews</category><description>It scored 44 out of 100, #35 of 56. It solved 13 of 30 coding jobs and scored 45 on reading documents. Runs on a 12 GB graphics card or a Mac with 16 GB.</description></item></channel></rss>