Gate News message, April 27 — SemiAnalysis, a semiconductor and AI analysis firm, released a comparative benchmark of coding assistants including GPT-5.5, Claude Opus 4.7, and DeepSeek V4. The key finding: GPT-5.5 marks OpenAI’s first return to the cutting edge in coding models in six months, with SemiAnalysis engineers now alternating between Codex and Claude Code after previously relying almost exclusively on Claude. GPT-5.5 is based on a new pretraining approach codenamed “Spud” and represents OpenAI’s first expansion of pretraining scale since GPT-4.5.
In practical testing, a clear division of labor emerged. Claude handles new project planning and initial setup, while Codex excels at reasoning-intensive bug fixes. Codex demonstrates stronger data structure comprehension and logical reasoning but struggles with inferring ambiguous user intent. On a single dashboard task, Claude automatically replicated the reference page layout but fabricated large amounts of data, while Codex skipped the layout but delivered significantly more accurate data.
The analysis reveals a benchmark manipulation detail: OpenAI’s February blog post urged the industry to adopt SWE-bench Pro as the new standard for coding benchmarks. However, GPT-5.5’s announcement switched to a new benchmark called “Expert-SWE.” The reason, buried in fine print, is that GPT-5.5 was surpassed by Opus 4.7 on SWE-bench Pro and fell significantly short of Anthropic’s unreleased Mythos (77.8%).
Regarding Opus 4.7, Anthropic published a postmortem analysis one week after release, acknowledging three bugs in Claude Code that persisted for several weeks from March to April, affecting nearly all users. Multiple engineers had previously reported performance degradation in version 4.6 but were dismissed as subjective observations. Additionally, Opus 4.7’s new tokenizer increases token usage by up to 35%, which Anthropic openly admitted—effectively constituting a hidden price increase.
DeepSeek V4 was assessed as “keeping pace with the frontier but not leading,” positioning itself as the lowest-cost alternative among closed-source models. The analysis also noted that “Claude continues to outperform DeepSeek V4 Pro on high-difficulty Chinese writing tasks,” commenting that “Claude won against the Chinese model in its own language.”
The article introduces a key concept: model pricing should be evaluated by “cost per task” rather than “cost per token.” GPT-5.5’s pricing is double that of GPT-5.4 (input $5, output $30 per million tokens), but it completes the same tasks using fewer tokens, making the actual cost not necessarily higher. Initial SemiAnalysis data shows Codex’s input-to-output ratio at 80:1, lower than Claude Code’s 100:1.
Disclaimer: The information on this page may come from third parties and does not represent the views or opinions of Gate. The content displayed on this page is for reference only and does not constitute any financial, investment, or legal advice. Gate does not guarantee the accuracy or completeness of the information and shall not be liable for any losses arising from the use of this information. Virtual asset investments carry high risks and are subject to significant price volatility. You may lose all of your invested principal. Please fully understand the relevant risks and make prudent decisions based on your own financial situation and risk tolerance. For details, please refer to
Disclaimer.
Related Articles
Should AI boost productivity or lower costs? A tenfold efficiency increase hasn’t turned into a tenfold revenue jump, but in Silicon Valley, nobody dares to call it off
Five Yuan Capital partner Meng Xing has recently published a Silicon Valley inspection report, proposing a judgment that has even changed his own note-taking habit: Silicon Valley is entering a stage where even people who can “ride waves” are drowned by the waves. The iteration speed of AI has shifted from “monthly” to “weekly”—even Silicon Valley itself can’t keep up with its own pace.
When AI amplifies a team’s productivity by five times, you can reduce 80% of the workforce to maintain the same output, or keep headcount and do five times the work. Meng Xing’s observations this time in Silicon Valley are essentially the first draft of the answer given on the ground: when 100x efficiency doesn’t translate into 100x revenue, when token budgets are edging toward human labor costs, and when the steam engine can’t outpace the horse carriage but no one dares to stop, Silicon Valley is choosing to “push speed up first and figure things out later.” But in the end, this path will lead to “expanding capability” or “compressing costs”—there’s currently no conclusion.
YC has gone from leading indicators to lagging indicators
Meng Xing this year
ChainNewsAbmedia6m ago
YC partners share how to use AI to build a company from scratch; startups should treat AI as an operating system rather than a tool
The impact of AI on startups is no longer limited to helping engineers write code faster, automating customer service workflows, or adding a Copilot to an existing product. Recently, YC partner Diana pointed out that the real change lies in the fact that AI is rewriting how a company should be built from scratch in the first place. For early founders, AI should not be merely an efficiency tool the company occasionally uses; it should be designed from day one as the operating system of the entire company.
The productivity perspective is outdated—AI is rewriting a company’s design starting point
Diana believes that when people in the market talk about AI today, they still too often stay within a “productivity improvement” framework—for example, engineers can write code faster, teams can automate more processes, and companies can roll out more features. But this argument actually underestimates the structural changes AI brings. She points out that the right people paired with AI…
ChainNewsAbmedia16m ago
Cursor AI agent caused an incident! One line of code cleared the company database in 9 seconds—“security checks” turned into empty talk
PocketOS founder Jer Crane said that Cursor AI agents ran maintenance on their own in a test environment, misused an API Token that adds/removes custom domains, and launched a delete command against Railway’s GraphQL API. Within 9 seconds, all data and same-region snapshots were completely destroyed, with the latest recoverable point being three months ago. The agent admitted to violating rules for irreversible operations, not reading technical documentation, not verifying environment isolation, and more. The victims were car rental industry customers; their bookings and all data disappeared, and reconciling accounts took a long time. Crane proposed five reforms: manual confirmation, fine-grained API permissions, backups separated from master data, a public SLA, and a mandatory underlying enforcement mechanism.
ChainNewsAbmedia18m ago
DeepSeek V4 Pro with Ollama Cloud: One-click integration with Claude Code
According to an Ollama tweet, DeepSeek V4 Pro was released on 4/24, has been added to the Ollama catalog in cloud mode, and can call tools like Claude Code, Hermes, OpenClaw, OpenCode, Codex, etc. with just a single line of command. V4 Pro: 1.6T params, 1M context, Mixture-of-Experts; cloud inference does not download local weights. If you want to run it locally, you need to obtain the weights yourself and run it with INT4/GGUF and multi-card GPUs. Early speed tests were affected by cloud load; typical performance is about 30 tok/s, with a peak of 1.1 tok/s. It is recommended to use the cloud prototype first, and for production later, run inference yourself or use a commercial API.
ChainNewsAbmedia1h ago
DeepSeek Cuts V4-Pro Prices by 75%, Slashes API Cache Costs to One-Tenth
Gate News message, April 27 — DeepSeek announced a 75% discount on its new V4-Pro model for developers and reduced input cache hit prices across its API lineup to one-tenth of previous levels.
The V4 model, released on April 25 in Pro and Flash versions, has been optimized for Huawei's Ascend
GateNews1h ago
Coachella turns to Google’s DeepMind AI to reimagine concerts beyond the stage
Coachella has partnered with Google DeepMind to test new AI tools that reshape how live music performances are created and experienced.
Summary
Coachella has tested AI tools with Google DeepMind to turn live performances into interactive digital environments.
Three prototypes were built,
Cryptonews1h ago
