{"slug": "how-to-run-groot-19b-a3b-261023-mtp-on-rtx-4070-super-rtx-4060-ti", "title": "How to run Groot-19B-A3B-261023 MTP on RTX 4070 SUPER + RTX 4060 Ti", "summary": "Groot-19B-A3B-261023 MTP Q5_K_M achieved perfect 30/30 needle recall across the full 262144-token served context window and the fastest spec-decode on the tested RTX 4070 SUPER plus RTX 4060 Ti setup, with embedded-MTP acceptance of 62-68%, but failed the compile gate 0/3 on the v5 benchmark. The three runs produced Elixir modules with defects including an inline @moduledoc heredoc at kv_1.ex:2:17, a semicolon-joined one-liner plus a Python-style raise call at kv_2.ex:13:73, and a do/end imbalance from an unless written without do at kv_3.ex:4:15, persisting at 33:5 after mechanical heredoc repair. The draft briefing concludes Groot is worth promoting to long-context retrieval and review duty behind a compile gate and a lint instruction to the orchestrator model, but is not a drop-in subagent model for Elixir coding.", "body_md": "**Warning:** this is a temporary briefing, written by an LLM and not yet evaluated or rewritten by a human to get rid of the signature AI-slop writing style. It’s published as a draft despite its shortcomings, so that you can find out how to run it and what to expect in terms of `pp` and `tg` on this hardware, as well as what to expect in terms of coding performance on this benchmark, *grosso modo*.\n\n**Groot-19B-A3B-261023 MTP Q5_K_M** demonstrated perfect (30/30) needle recall across the full served 262144-token context window, the fastest spec-decode among the models measured on this box (embedded-MTP acceptance 62-68%), and proper ```` ``` elixir ```` fences in all three runs, perfectly fulfilling the output contract of the [v5 benchmark](https://overbring.com/benchmark-v5.en). However, it still missed the compile gate 0/3 with more issues than merely a typo: the modules mix Python block syntax (unless cond with no do) with heredoc and one-liner issues, and run 3 keeps failing even after the heredocs are mechanically repaired. Review quality is the best measured on this hardware so far, but code output is not ready to compile.\n\n## Representative output\n\nRun 3 of the benchmark delivered the cleanest, most complete module of the batch, with full `@spec` coverage on the public API and a timer-driven eviction loop. Trimmed excerpt of the public API:\n\n``` python\ndefmodule TtlStore do\n  @spec start_link(map()) :: pid\n  def start_link(opts) when is_map(opts) do\n    ttl = Map.get(opts, :ttl, 3600)\n    validate_ttl!(ttl)\n    state = make_state(ttl)\n    {pid, _ref} = Process.spawn_linker(fn s -> loop(s) end, state)\n    pid\n  end\n\n  @spec start_link(binary(), integer()) :: pid\n  def start_link(name, ttl) when is_binary(name), is_integer(ttl) do\n    start_link(%{name: name, ttl: ttl})\n  end\n\n  @spec put(pid, binary(), binary()) :: atom()\n  def put(pid, key, value) do\n    validate_key!(key)\n    validate_value!(value)\n    send_cmd(pid, {:put, key, value})\n  end\n\n  @spec put(pid, binary(), binary(), integer()) :: atom()\n  def put(pid, key, value, ttl) do\n    validate_key!(key)\n    validate_value!(value)\n    validate_ttl!(ttl)\n    send_cmd(pid, {:put, key, value, ttl})\n  end\n\n  @spec get(pid, binary()) :: {atom(), binary()}\n  def get(pid, key) do\n    validate_key!(key)\n    send_cmd(pid, {:get, key})\n  end\n\n  @spec delete(pid, binary()) :: atom()\n  def delete(pid, key) do\n    validate_key!(key)\n    send_cmd(pid, {:delete, key})\n  end\n\n  @spec stats(pid) :: map()\n  def stats(pid) when is_pid(pid) do\n    send_cmd(pid, :stats)\n  end\nend\n```\n\n## The catch\n\nThe delivered-pipeline gate is 0/3 as-filed, with three distinct issues, each one in its own defect class:\n\n- run 1 `kv_1.ex:2:17` — inline`@moduledoc` heredoc (triple-quote content on the opening line).\n- run 2 `kv_2.ex:13:73` — a ;-joined one-liner`def f(_x); … end` plus the Python`raise(ArgumentError, …)` call-shape.\n- run 3 `kv_3.ex:4:15` → post-patch`33:5` — inline heredoc, then a do/end imbalance (`unless cond` written without`do` ) that survives a mechanical heredoc repair.\n\nThe issue runs two levels deep: cosmetic tokens **and** real block structure. Format discipline held across all three runs (proper ```` ``` elixir ```` fences), fulfilling the benchmark prompt’s output contract). However, the model’s output requires genuine editing by the orchestrator model that goes beyond small changes.\n\n## Route\n\nIt’s worth promoting Groot to long-context retrieval and review duty, but keep a compile gate and a “lint your output” instruction to the orchestrator in front of it. This is not a drop-in subagent model for Elixir coding that will require no orchestration, and can therefore not be used as a primary model for that purpose either.", "url": "https://wpnews.pro/news/how-to-run-groot-19b-a3b-261023-mtp-on-rtx-4070-super-rtx-4060-ti", "canonical_source": "https://overbring.com/ai-benchmarks/2026-10-31-prithivmlmods-groot-19b-a3b-261023-q5-k-m-gguf-rtx-4070-super-12gb-plus-4060-ti-16gb-layer-split-llama-server-cuda/", "published_at": "2026-10-31 21:23:57+00:00", "updated_at": "2026-11-09 01:58:29.078372+00:00", "lang": "en", "topics": ["large-language-models", "ai-tools", "ai-infrastructure"], "entities": ["Groot-19B-A3B-261023 MTP", "RTX 4070 SUPER", "RTX 4060 Ti", "Elixir"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/how-to-run-groot-19b-a3b-261023-mtp-on-rtx-4070-super-rtx-4060-ti", "markdown": "https://wpnews.pro/news/how-to-run-groot-19b-a3b-261023-mtp-on-rtx-4070-super-rtx-4060-ti.md", "text": "https://wpnews.pro/news/how-to-run-groot-19b-a3b-261023-mtp-on-rtx-4070-super-rtx-4060-ti.txt", "jsonld": "https://wpnews.pro/news/how-to-run-groot-19b-a3b-261023-mtp-on-rtx-4070-super-rtx-4060-ti.jsonld"}}