From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mta0.migadu.com (out-110.mta0.migadu.com [91.218.175.110]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 69AB2470105 for ; Thu, 10 Sep 2026 11:08:59 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=91.218.175.110 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1789038544; cv=none; b=svKJabN1yb14coegNMR+/IkmPvv+/5TnsKSGEdL3LLx9aUVcHloQCoPq5CTzLBkp3oEQF2Et0ZO6y7Znc8OuFjlpv+Go78K6ZsLRP7oAgNexO5QnepB6dKJ7zcgkECyPiQGekS5C4867PIIPT7eYF9Cn+kVdWSRaDJt90CtX2D8= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1789038544; c=relaxed/simple; bh=YZJKBp1Xhz1wtl66ZjSk3W6Iyi0VyQ8+Z80goYlTkCM=; h=MIME-Version:Date:Content-Type:From:Message-ID:Subject:To: In-Reply-To:References; b=BEg2hWzSeys2FisWQGSnmZT7kPxTT+e+lCFdT2pZ1Mtrd2OfFXEjPefI6ImQ4cmX9pfjynjKwsQO6sgkGyZ24VySO2cztvr24tbpdMdradWAo+2mj14MX/IMKQqjmlecgPZShzzXNo7COeQu/UHx+0lsgOkuUPmsWEu5hLlMpW4= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.dev; spf=pass smtp.mailfrom=linux.dev; dkim=pass (1024-bit key) header.d=linux.dev header.i=@linux.dev header.b=fEK4HRs/; arc=none smtp.client-ip=91.218.175.110 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.dev Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=linux.dev Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linux.dev header.i=@linux.dev header.b="fEK4HRs/" X-Envelope-To: mptcp@lists.linux.dev DKIM-Signature: a=rsa-sha256; bh=YZJKBp1Xhz1wtl66ZjSk3W6Iyi0VyQ8+Z80goYlTkCM=; c=simple/simple; d=linux.dev; h=from:to:subject:date:message-id:mime-version:content-type; s=key1; t=1789038537; v=1; x=1789643337; b=fEK4HRs/yTIT7//J1dsC0bxNJ/3VVw4FsRsIzK7Vi8pHj0MQ8USSLiz4qdP8Qif2dHqSNBAW ux6xCVIzhPJNqAVk2qWRk9MSoliSeShIUoN6TAz/ArbRvxKlmHzIV0HIYj/5NTBwcMVqUGbzFIw mqkJ4ir0GFVor9zKiS1eUVKU= X-Envelope-To: mptcp@lists.linux.dev Received: by smtp.migadu.com with ESMTPS id 327d5b8c6aa10446; Thu, 10 Sep 2026 11:08:57 +0000 X-Mizu-Trace-ID: 327d5b8c6aa10446 X-Migadu-Flow: FLOW_OUT Precedence: bulk X-Mailing-List: mptcp@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Date: Thu, 10 Sep 2026 11:08:57 +0000 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable From: gang.yan@linux.dev Message-ID: <703f2d892ac7432d4317f949e6ceba1b7f73ba40@linux.dev> TLS-Required: No Subject: Re: [RFC] Mitigating Sashiko-driven iteration noise on the main MPTCP list To: "Shardul Bankar" , mptcp@lists.linux.dev In-Reply-To: <20260907125720.516598-1-shardul.b@mpiricsoftware.com> References: <6f86381a-72d4-4a47-bc7d-1880cc39da64@kernel.org> <3f13f3acd4c329480331cccf5a12411dc01ee721@linux.dev> <7429a238-6297-4a7c-b18c-47cbcee4ab69@kernel.org> <20260907125720.516598-1-shardul.b@mpiricsoftware.com> >=20 >=20Gang, you mentioned deploying Sashiko locally. Did you see anything l= ike > the same cost, and what were you running it against? If this turns out > to be particular to a fixed-quota subscription, I would rather find tha= t > out than have people avoid running reviews locally over a cost that doe= s > not apply to them. >=20 Hi=20Shardul, Thanks for your reply. Sorry for the delayed reply; work has kept me busy lately so I haven=E2= =80=99t had much time to check emails. Please also excuse my English =E2=80=94 I might hav= e caused some confusion earlier. I'm using the API key and noticed the cost. I=E2=80=99d like to share my experience with an API-key Sashiko deploymen= t, which may help folks on this list use Sashiko at lower cost. I run Sashiko using Minimax as the LLM. Minimax offers a monthly subscrip= tion plan and is quite affordable. (I submitted a small PR to Sashiko to fix some i= ssues I encountered while using it.) One observation: reviewing large patch series via the web interface can b= urn through the 5-hour quota very quickly. In contrast, using `sashiko-cli` from the = command line seems to cut token consumption significantly, this is from my practical t= esting. I also tried GLM, which is another fixed-quota subscription model and rel= atively more expensive. While its quality is not as good as Gemini or Opus, it st= ill works well for local code iteration. I=E2=80=99m still relatively new to working with AI tools. If you have an= y approaches to save tokens, would you mind sharing them on this thread? Thanks Gang > Thanks, > Shardul >