{"slug": "evaluating-muse-spark-1-3-on-hack-the-box-challenges", "title": "Evaluating Muse Spark 1.3 on Hack The Box Challenges", "summary": "Meta's Muse Spark 1.3 scored 72.93% on the HTB-Challenger Benchmark, up from 50.41% for Muse Spark 1.2, according to a blog evaluation of the model on Hack The Box security challenges. The new version solved 14 of 16 challenges at a median cost of $0.38 per challenge, but the reviewer said there is no reason to switch from OpenAI's GPT-5.6 Sol (87.2% / $0.29) or Z.ai's GLM 5.3 Flash (81.0% / $0.01). Meta's announcement claimed Muse Spark 1.3 uses approximately 20% fewer tool calls and 25% fewer tokens than Muse Spark 1.2.", "body_md": "# Evaluating Muse Spark 1.3 on Hack The Box Challenges\n\nThe company-who-must-not-be-named released a new version of its Muse Spark model - **Muse Spark 1.3**. The [announcement claims](https://research.meta.ai/blog/introducing-muse-spark-1-3) that this model should be more suitable for long-horizon tasks and should use approximately 20% fewer tool calls and 25% fewer tokens than **Muse Spark 1.2**. I [tested the 1.2 version](https://theaq.blog/2026/08/17/solving-htb-challenges-with-meta-muse-spark-1.2.html) a few weeks ago, and I found it forgettable. So I was curious whether a gap of just one month between the two model releases could make any difference.\n\nThis blog post is part of a series of tests for the [HTB-Challenger Benchmark](https://theaq.blog/htb-challenger-benchmark/)The HTB-Challenger Benchmark evaluates LLMs’ ability to find and exploit security vulnerabilities. It tests models against selected Hack The Box challenges of varying difficulty and measures their performance. For more information, visit the [HTB-Challenger Benchmark page](https://theaq.blog/htb-challenger-benchmark/).\n.\nSee the [benchmark results page](https://theaq.blog/htb-challenger-benchmark) for all results and the [benchmark methodology](https://theaq.blog/htb-challenger-benchmark-methodology) to learn how the benchmark is calculated.\n\nSurprisingly, it turns out that you can do a lot in four weeks in terms of improving an LLM for solving HTB challenges. The new version achieved a much better score than its predecessor: 72.93% versus 50.41%. Its score is in the same range as those of [Kimi K3](https://theaq.blog/2026/08/10/solving-htb-challenges-with-moonshotai-kimi-k3.html) (70.67%) and [Grok 4.6](https://theaq.blog/2026/08/12/solving-htb-challenges-with-x-ai-grok-4.6.html) (76.14%), both of which are great models.\n\nThe only issue is the median cost per challenge, which is better than the previous version ($0.476 vs $0.385), but still relatively high. You can use [GPT-5.6 Sol](https://theaq.blog/2026/08/16/solving-htb-challenges-with-openai-gpt-5.6-sol.html), which has a much better benchmark score and a slightly lower cost (87.2% / $0.29), or my favorite, [GLM 5.3 Flash](https://theaq.blog/2026/08/28/evaluating-z-ai-glm-5.3-flash-on-hack-the-box-challenges.html), with a slightly higher score and a much lower cost (81.0% / $0.01).\n\nSo the new **Muse Spark 1.3** is not bad at all, but I don’t see any reason to switch to it from **GPT-5.6 Sol** or **GLM 5.3 Flash**.\n\n## Cost vs. Benchmark Score\n\nThe highlighted point is this model. Models closer to the upper-left achieve a higher benchmark score at a lower median cost per challenge.\n\n## Overall benchmark results\n\n- **Number of challenges:** 16\n- **Number of solved challenges:** 14\n- **Number of false positives:** 0\n- **Runs where the model gave up:** 1\n- **Runs that reached the step or cost limit:** 1\n- **Runs where the model got stuck:** 0\n- **Benchmark score:** 72.9%\n\n| Metric | Per challenge (median) | Total | \n|---|---|---|\n| Model steps | 23 | 608 | \n| Model cost | $0.38 | $13.64 | \n| Duration | 00:08:05 | 02:41:53 | \n| Number of input tokens | 0.60M | 23.44M | \n| Number of output tokens | 0.02M | 0.58M | \n| Number of `read_file` tool calls | 1.0 | 64 | \n| Number of `write_file` tool calls | 0.0 | 15 | \n| Number of `execute_command` tool calls | 19.5 | 524 | \n| Number of `web_search` tool calls | 0.5 | 25 | \n\n## Results by challenge difficulty\n\nAll resource-usage metrics are medians per challenge.\n\n| Metric | Very Easy | Easy | Medium | Hard | \n|---|---|---|---|---|\n| **Results** |  |  |  |  | \n| Number of challenges | 4 | 4 | 4 | 4 | \n| Number of solved challenges | 4 | 4 | 4 | 2 | \n| Number of false positives | 0 | 0 | 0 | 0 | \n| Runs where the model gave up | 0 | 0 | 0 | 1 | \n| Runs that reached the step or cost limit | 0 | 0 | 0 | 1 | \n| Runs where the model got stuck | 0 | 0 | 0 | 0 | \n| Benchmark score | 93.7% | 97.8% | 88.4% | 43.7% | \n| **Median per challenge** |  |  |  |  | \n| Model steps | 18 | 11.5 | 42 | 73.5 | \n| Model cost | $0.17 | $0.11 | $0.98 | $1.87 | \n| Duration | 00:02:02 | 00:04:01 | 00:10:46 | 00:17:39 | \n| Number of input tokens | 0.32M | 0.12M | 1.58M | 3.18M | \n| Number of output tokens | 0.01M | 0.01M | 0.04M | 0.06M | \n| Number of `read_file` tool calls | 0.5 | 1.0 | 2.0 | 3.5 | \n| Number of `write_file` tool calls | 0.0 | 0.0 | 0.0 | 0.5 | \n| Number of `execute_command` tool calls | 13.5 | 8.0 | 39.0 | 61.5 | \n| Number of `web_search` tool calls | 0.5 | 1.5 | 0.0 | 2.5 |", "url": "https://wpnews.pro/news/evaluating-muse-spark-1-3-on-hack-the-box-challenges", "canonical_source": "https://theaq.blog/2026/09/06/evaluating-meta-muse-spark-1.3-on-hack-the-box-challenges.html", "published_at": "2026-09-06 09:15:41+00:00", "updated_at": "2026-09-13 18:22:36.651490+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "ai-agents", "ai-research"], "entities": ["Meta", "Muse Spark 1.3", "Muse Spark 1.2", "HTB-Challenger Benchmark", "Hack The Box", "GPT-5.6 Sol", "GLM 5.3 Flash", "Kimi K3"], "alternates": {"html": "https://wpnews.pro/news/evaluating-muse-spark-1-3-on-hack-the-box-challenges", "markdown": "https://wpnews.pro/news/evaluating-muse-spark-1-3-on-hack-the-box-challenges.md", "text": "https://wpnews.pro/news/evaluating-muse-spark-1-3-on-hack-the-box-challenges.txt", "jsonld": "https://wpnews.pro/news/evaluating-muse-spark-1-3-on-hack-the-box-challenges.jsonld"}}