AI News

Trending story

Google launches Gemini 4 Hydrogen on Antigravity

49 reports3 discussions52 story updatesUpdated Nov 15 09:45

Understand this story

AI overview

Google DeepMind released Gemini 4 Hydrogen, emphasizing speed and efficiency for everyday coding and agentic workflows, and simultaneously launched Orchestrator in Google Antigravity, which can dynamically switch models between Carbon and Hydrogen within the same session. The company says the model scores 51 on the Artificial Analysis Intelligence Index at the default high setting and 54 at the xhigh setting, has a 1 million token context window, is priced at $1.50 per million input tokens and $7.50 per million output tokens, and offers four thinking levels: low, medium, high, and xhigh. Google subsequently integrated Gemini 4 Hydrogen into Google Search and made it available to all users worldwide, calling it the largest model upgrade to Search to date. Artificial Analysis benchmarks show it scores 54 at the xhigh setting, 1 point lower than Claude Sonnet 5.5 (max) and GPT-6.1 Sol (max); Arena.ai calls it the fastest model in the Agent Arena top ten. Cursor announced that Gemini 4 Hydrogen is now available in Cursor, scoring 54.3% on CursorBench at the default effort. S

AI-generated from public reports

Latest development · Report

Edge AI Daily Morning Brief (November 15): NVIDIA's Jensen Huang Resigns as CEO; Google Releases Gemini 4 Hydrogen

Developments

7 confirmed developments

  1. Gemini 4 Hydrogen (Flash) launches on Antigravity

    Source:Google DeepMind
  2. Artificial Analysis evaluates Gemini 4 Hydrogen

    Source:Artificial Analysis
  3. Arena.ai calls Gemini 4 Hydrogen the fastest model in the Agent Arena top ten

    Source:Arena
  4. Cursor launches Gemini 4 Hydrogen

    Source:Cursor
  5. Cursor launches Gemini 4 Hydrogen

    Source:Cursor
  6. Simon Willison releases llm-gemini 0.37

    Source:Simon Willison
  7. Google brings Gemini 4 Hydrogen to Google Search

    Source:Google Blog

Reports and discussion timeline

Browse reports and discussion by time. Their evidence levels remain distinct.

Nov 15

  1. TMTPost · AGI

    Edge AI Daily Morning Brief (November 15): NVIDIA's Jensen Huang Resigns as CEO; Google Releases Gemini 4 Hydrogen

    On November 12, NVIDIA announced that Jensen Huang has resigned from his roles as President, CEO, and board member, effective immediately; CFO Colette Kress has taken over as interim CEO, and the board has begun the search for a successor. Signatures on the petition website have exceeded 115,000. Google released its execution-type model Gemini 4 Hydrogen the same day, supporting four thinking levels and a 1 million Token context, with API pricing at $1.50 per million input Tokens and $7.50 per million output Tokens. This morning brief also covers the FAA clearing Starship's 15th flight, Microsoft 365 being down for more than 24 hours, and other news.

Nov 14

  1. cnBetaSelected

    Google Releases Gemini 4 Hydrogen and Integrates It into Search

    Google has released Gemini 4 Hydrogen, which is now available to users worldwide and covers all languages supported by AI Mode. Google says that at the default High reasoning level, it completes tasks 2.5 times faster than the previous-generation Gemini 3.8 Flash, with more than 80% fewer fabricated facts; on the Artificial Analysis Intelligence Index, it scores 51 at the High level and 54 at the Xhigh level, surpassing Gemini 4 Argon's 50. API pricing is $1.5 per million input tokens and $7.5 per million output tokens, lower than Argon's $2 and $10; Google has also launched Orchestrator for Antigravity, which can automatically switch between Carbon and Hydrogen within the same task, initially available to Google AI Ultra users.

  2. cnBetaRelated discussion

    cnBeta joined this discussion

    This post contributes to heat; it is not a formal report or confirmed fact.

  3. TMTPost · AGI

    Edge AI Daily Morning Brief: Nvidia's Jensen Huang Resigns, Anthropic's $90.4 Billion IPO and Google…

    Nvidia announced that Jensen Huang is immediately stepping down as President, CEO and board director, with CFO Colette Kress serving as interim CEO; on the day of the news, the stock fell about 14% in early trading and its market value dropped below $4 trillion. Anthropic's IPO attracted about six times subscription, with the public offering raising up to $64.4 billion and accompanied by a $26 billion private placement, for a combined total of possibly $90.4 billion. Google released Gemini 4 Hydrogen, which is 2.5 times faster than Gemini 3.8 Flash, with API pricing at $1.50 per million input tokens and $7.50 for output.

  4. Arena

    Arena Weekly: Claude Fable 5.2 Tops Agent Arena, Gemini 4…

    This week's Arena leaderboard shows that Anthropic's Claude Fable 5.2 (Max) tops Agent Arena, achieving a +30.5% net improvement based on 7.8k real sessions. Google DeepMind's Gemini 4 Hydrogen (High) is the fastest model in the Agent Arena top ten, with a typical single-step latency of 7.3 seconds. Meta's smallest model, Muse Ember (Max), entered Code Arena: WebDev for the first time, ranking 37th with 1609 points and a median cost of $0.03 per task; thinkymachines' Inkling 2 Small, after opening its weights, became the top US open-source model on that leaderboard with 1647 points; MiniMax H3-Image ranked first among open-source models and 11th overall on Multi-Image Edit with 1365 points; Baidu's ERNIE 5.5 Flash Preview, with 1668 points (AutoEval), is closing in on second place among open-source models on that leaderboard, with a 1M token context window.

  5. ThurrottRelated discussion

    Thurrott joined this discussion

    This post contributes to heat; it is not a formal report or confirmed fact.

Nov 13

  1. 36Kr · AISelected

    Google Releases Gemini 4 Hydrogen and Sets It as the Default Model for Search AI Mode

    On November 12, Google released Gemini 4 Hydrogen and rolled it out across the Gemini app, AI Studio, API, Gemini for Business, and Antigravity, while also making it the default model for Google Search AI Mode. Third-party Artificial Analysis gives it 51 points on the default high tier, above its own flagship Argon's 50 points, with the xhigh tier reaching 54 points; however, Terminal-Bench 4.1 is 57%, below Claude Sonnet 5.5's 63.9%. API pricing is $1.5 per million input tokens and $7.5 per million output tokens, double that of 3.8 Flash's 0.75/3.75. The Orchestrator launched alongside it can dynamically switch between Carbon and Hydrogen within the same session, and Google says it consumes 24% less quota and is 19% faster.

  2. English text is being prepared.Selected

    Google Releases Gemini 4 Hydrogen Model, Focused on Everyday Coding and Agentic Workflows

    Google has officially launched the Gemini 4 Hydrogen model, aimed at everyday coding and agentic workflows, priced at $1.50 per million input Tokens and $7.50 per million output Tokens, covering a 1 million Token context window with no tiered pricing. The model offers four thinking depth levels: low, medium, high (default), and xhigh. On the Artificial Analysis Intelligence Index, it scores 51 at the default high thinking level and can reach 54 at the xhigh level. Google has also simultaneously launched the Orchestrator intelligent scheduling feature on the Google Antigravity platform, which can automatically switch between models such as Carbon and Hydrogen within the same session; in Terminal-Bench 4.0 tests, average quota consumption was reduced by 24% and completion speed improved by 19%.

  3. IT Home · AIRelated discussion

    IT Home · AI joined this discussion

    This post contributes to heat; it is not a formal report or confirmed fact.

  4. Analytics India Magazine

    Google Releases Gemini 4 Hydrogen, Focused on Coding and AI Workflows

    Google has released Gemini 4 Hydrogen, aimed at coding, research, and everyday business tasks. It belongs to the Gemini 4 family, which also includes Gemini 4 Carbon and Gemini 4 Argon. Google says Hydrogen scores 51 on the Artificial Analysis Intelligence Index at the default high thinking level, rising to 54 at the highest level; on Terminal-Bench 4.1 it scores 57%, compared with 18.2% for Gemini 3.8 Flash. The model has a context window of 1 million tokens, shared between input and output, and is priced at $1.50 per million input tokens and $7.50 per million output tokens. It is being rolled out through the Gemini API, Google AI Studio, and other channels.

  5. heise online

    Google releases Gemini 4 Hydrogen, focusing on speed and cost efficiency

    Google has released Gemini 4 Hydrogen, the third model in the Gemini 4 family. Derived from Carbon, it focuses on speed and cost efficiency and replaces Gemini 3.8 Flash. It supports text, image, audio, and video input, has a context window of one million tokens, outputs text only, and offers four thinking modes; on Artificial Analysis's Intelligence Index, its highest xhigh mode scores 54 points, slightly below the 55 points of Claude Sonnet 5.5 and GPT-6.1 Sol. API pricing is $1.50 per million input tokens and $7.50 per million output tokens, and it has launched with Google Search, the Gemini app, Google AI Studio, and the API.

  6. YicaiSelected

    Google Releases Lightweight Model Gemini 4 Hydrogen, Launching Three Major Models in Six Weeks

    Beijing time November 13, Google released the new model Gemini 4 Hydrogen, the third model it has released within six weeks, with a focus on faster response, lower cost, and agent execution capabilities. Hydrogen is open to users worldwide, covering the Gemini app, Google Search, AI Studio, API, and enterprise services, and free users can also use it through the automatic model selection feature. Its API pricing is $1.5 per million input tokens and $7.5 per million output tokens; on the Terminal-Bench 4.1 test, its score rose from 18.2% to 57%. Google also launched Orchestrator in its AI coding platform Google Antigravity, supporting dynamic switching between different models within the same task.

  7. Sina Tech

    Google Releases Lightweight Model Gemini 4 Hydrogen, Its Third Major Model in Six Weeks

    On November 13, Google released the new model Gemini 4 Hydrogen, its third model release within six weeks, with a focus on faster responses, lower costs, and more efficient agent execution capabilities. The model is available to users worldwide, covering the Gemini app, Google Search, AI Studio, the API, and enterprise services, and free users can access it through the automatic model selection feature. API pricing is $1.5 per million input tokens and $7.5 per million output tokens; on the Terminal-Bench 4.1 test, its score rose from 18.2% for the previous generation Gemini 3.8 Flash to 57%. Google also launched Orchestrator in its AI coding platform Google Antigravity, supporting dynamic switching between different models within the same task.

  8. 36Kr · AISelected

    Google Releases Gemini 4 Hydrogen, 2.5x Faster and Focused on Low Cost

    Google has released Gemini 4 Hydrogen, positioned as a fast-execution model for everyday coding and agent tasks, competing with Claude Sonnet 5.5 and GPT-6.1 Terra. Google says it scores 13.5% on SWE-sweep, 71.6% on Agents' Last Exam v1.1, and 76.1% on AutomationBench-AA, all higher than the two competitors; it trails on OSWorld 2.1 and Terminal-Bench-Science 0.1. Artificial Analysis data shows its top-tier score is 54, close to Sonnet 5.5's 55, with an average cost of $3.16 per task, about one-third of Sonnet 5.5's $10.02. The model is being gradually rolled out to the Gemini app, Google Search, AI Studio, and the Gemini API.

  9. gihyo.jpSelected

    Google releases Gemini 4 Hydrogen and launches automatic model switching in Antigravity…

    On November 12, 2026, Google released Gemini 4 Hydrogen, aimed at everyday coding and multi-step agentic tasks, and made it available progressively the same day through channels including the Gemini API, the Gemini app, and Google Search. The model scores 57% on Terminal-Bench 4.1, higher than Gemini 3.8 Flash's 18.2% but lower than Claude Sonnet 5.5's 63.9%. The Antigravity development environment simultaneously introduced Orchestrator, which can automatically switch between Carbon and Hydrogen within the same session based on the task; the API model ID is gemini-4-hydrogen, and thinking_level has four tiers—low, medium, high, and xhigh—with high as the default.

  10. Phoenix TechnologySelected

    Google Releases Gemini 4 Hydrogen Model and Launches Orchestrator Scheduling System

    On November 12, Google officially released Gemini 4 Hydrogen, the successor to Gemini 3.8 Flash. It is natively multimodal, with input, thinking, and output sharing a 1 million token window. On the Artificial Analysis Intelligence Index, it scores 51 points in the default high tier and 54 points in the xhigh tier, surpassing the larger-parameter Argon (50 points). API pricing is $1.5 per million input tokens and $7.5 per million output tokens. Google also launched the multi-model scheduling system Orchestrator, which is already available to Ultra users in Google Antigravity, and uses Hydrogen as the default model for AI Mode in its search engine.

  11. TMTPost · AGISelected

    Google Releases Gemini 4 Hydrogen, Opens It to All Users and Integrates It into Search

    Google has launched Gemini 4 Hydrogen, which focuses on speed and execution efficiency. Free users can use it through the automatic model selection in the Gemini App, paid users can select it manually, and developers can call it through Google AI Studio and the Gemini API. Google says that at the default High reasoning level, it completes tasks 2.5 times faster than Gemini 3.8 Flash and reduces fabricated facts by more than 80%. On the Artificial Analysis Intelligence Index, it scores 51 at the High level and 54 at the Xhigh level, surpassing Gemini 4 Argon's 50. Google Search has also integrated the model and upgraded the Query fan-out query expansion technology; API pricing is $1.5 per million input tokens and $7.5 per million output tokens.

  12. 36Kr · AI

    Google Releases Gemini 4 Hydrogen, Open to All Users

    Google has launched Gemini 4 Hydrogen, which focuses on speed and execution efficiency. Free users can use it through automatic model selection in the Gemini App, paid users can select it manually, and developers can access it via Google AI Studio and the Gemini API. Google says that at the default High reasoning level, it completes tasks 2.5 times faster than the previous generation Gemini 3.8 Flash, and fabricates facts more than 80% less often. On the Artificial Analysis Intelligence Index, Hydrogen scores 51 at the High level and 54 at the Xhigh level, surpassing Gemini 4 Argon's 50; API pricing is $1.5 per million input tokens and $7.5 per million output tokens. Google Search has also integrated Hydrogen, and Google says this is the largest model upgrade to its search to date.

  13. sspai

    Paid Morning Brief: NVIDIA CEO Jensen Huang Announces Resignation; Volcano Engine Launches Seedance 2.5 flash and More

    NVIDIA's board of directors announced that co-founder Jensen Huang has resigned from his roles as President, CEO, and director, effective immediately, with CFO Colette Kress serving as interim CEO—the first CEO change since NVIDIA's founding. On the same day, Volcano Engine launched the video model Seedance 2.5 flash, doubling generation speed and cutting prices by about 34%, with 720P text-to-video at approximately 0.99 yuan per second; the API is scheduled to go live on Volcano Ark on November 15. Google also released Gemini 4 Hydrogen, supporting a 1 million Token context, with API input at $1.50 per million Tokens.

  14. Phoenix Technology

    Jensen Huang Resigns as NVIDIA CEO; Google Releases Gemini 4 Hydrogen | AI Morning Express

    On November 12, NVIDIA announced that founder Jensen Huang has resigned from his roles as President, CEO, and Director, effective immediately; Chief Financial Officer Colette Kress will concurrently serve as interim CEO, and the board has initiated a search for a successor. On the same day, Google launched Gemini 4 Hydrogen, focused on everyday coding and multi-step agentic tasks, capable of processing text, images, audio, and video, with API pricing at $1.5 per million input tokens and $7.5 per million output tokens. Several U.S. public interest organizations filed suit in San Francisco against OpenAI, Google, Anthropic, Meta, and SpaceX's AI business, asking the court to block them from continuing to develop advanced AI models.

  15. Phoenix TechnologySelected

    Google Releases Gemini 4 Hydrogen Model, Available to Free Users Too

    Google has released Hydrogen, a new model in the Gemini 4 series, featuring faster responses, lower cost, and stronger AI Agent execution capabilities, and it is integrated into platforms including the Gemini App, Google Search, Google AI Studio, and Antigravity; free users of the Gemini App can also use it through model selection. Artificial Analysis evaluations show that Hydrogen scores 51 on the intelligence index in the default High tier, surpassing Gemini 4 Argon's 50, and rises to 54 when the newly added xhigh tier is enabled; API pricing is $1.5 per million input tokens and $7.5 per million output tokens, with support for a 1 million token context window. Google has also launched the Orchestrator feature for Antigravity, which can automatically switch between models such as Carbon and Hydrogen within the same session based on the task; the company says it maintains success rates close to Carbon on Terminal-Bench 4.0 while reducing quota consumption by an average of 24% and increasing task completion speed by 19%.

  16. Phoenix Tech

    Google Releases Gemini 4 Hydrogen, Open to All Users

    Google has launched Gemini 4 Hydrogen, which prioritizes speed and execution efficiency. Free users can use it through automatic model selection in the Gemini App, paying users can select it manually, and developers can access it via Google AI Studio and the Gemini API. Google says that at the default High reasoning level, it completes tasks 2.5 times faster than the previous generation Gemini 3.8 Flash, with fabricated facts reduced by more than 80%. On the Artificial Analysis Intelligence Index, Hydrogen scores 51 at the High level and 54 at the Xhigh level, surpassing Gemini 4 Argon's 50; API pricing is $1.5 per million input tokens and $7.5 per million output tokens.

  17. Sina TechSelected

    Google Releases Gemini 4 Hydrogen Model with 2.5x Speed Improvement

    Google has officially released the Gemini 4 Hydrogen model, focusing on efficient coding and AI agent tasks. Google says that in the default High thinking mode, its task completion speed is 2.5 times that of Gemini 3.8 Flash, and it scores 57% on the Terminal-Bench 4.1 coding benchmark, compared with 18.2% for the previous generation. The model supports a context window of up to 1 million tokens and has been launched globally through the Gemini app, Google Search, Google AI Studio, and the Gemini API. API pricing is $1.5 per million input tokens and $7.5 per million output tokens.

  18. Simon Willison

    llm-gemini 0.37 released, with support for Gemini 4 Hydrogen

    Simon Willison released llm-gemini 0.37, and this LLM plugin adds support for Gemini 4 Hydrogen, which Google launched the same day. Hydrogen is priced at $1.50/$0.075/$7.50, currently only 25% cheaper than Carbon and Argon, while those two will rise on December 1 from $2.00/$0.10/$10.00 to $4.00/$0.20/$20.00. The author used the three models at three effort levels to generate pelican images for comparison, concluded that the image-quality gap between Hydrogen and its same-price sibling models is obvious, and noted that Hydrogen is the new free-tier model and will also power AI features in Google Search.

  19. SiliconANGLESelected

    Google releases Gemini 4 Hydrogen, claims it surpasses Argon on an independent intelligence index

    Google has released Gemini 4 Hydrogen, the lower-cost model in the Gemini 4 family, claiming that it scores 51 on the Artificial Analysis Intelligence Index at the default high thinking setting, above Gemini 4 Argon's 50, and that the newly added xhigh setting can reach 54. The model skips the nearly two-week early-access phase that Argon had for Fairwind Program security personnel, and is available to all users starting today, including users of the free version of the Gemini app. Google says Hydrogen scores 57% on Terminal-Bench 4.1, compared with 18.2% for Gemini 3.8 Flash; the Orchestrator added to the Antigravity coding tool the same day switches between Hydrogen and Gemini 4 Carbon, approaching Carbon's success rate on Terminal-Bench 4.0 while using an average of 24% less quota.

  20. Phoenix TechnologySelected

    Google Releases Gemini 4 Hydrogen, Free Users Can Access It via Automatic Model Selection

    Google has released its new-generation AI model Gemini 4 Hydrogen, focusing on improving the speed and efficiency of everyday coding and AI agent task execution. It has begun rolling out to users worldwide, and free users of the Gemini app can access it through the automatic model selection feature. Google says Hydrogen focuses on quickly completing everyday coding, tool calling, and multi-step agent tasks, forming a division of labor with Gemini 4 Carbon and Argon, which target difficult, long-reasoning tasks. The model offers four thinking levels: low, medium, high, and highest. API pricing is $1.5 per million input tokens and $7.5 per million output tokens, with a context window of 1 million tokens, and developers can use it through Google AI Studio and the Gemini API.

  21. cnBetaSelected

    Google releases Gemini 4 Hydrogen, focusing on speed and everyday agent tasks

    Google has officially released Gemini 4 Hydrogen, positioned as a fast model for handling everyday work and agent tasks, with native support for 1 million Token input and up to 1 million Token output. Google says it scores 57% on Terminal-Bench 4.1, compared with 18.2% for Gemini 3.8 Flash; on the Artificial Analysis Intelligence Index it scores 51 under the default High reasoning setting and 54 under the XHigh setting. API pricing is $1.50 per million Tokens for input and $7.50 for output. It is now available to users at all tiers of Google AI Studio, Gemini API developers, Gemini for Business enterprise users, and Gemini app users, and is integrated into Google Search.

  22. The DecoderSelected

    Google releases Gemini 4 Hydrogen, with the default thinking level at Artificial…

    Google releases Gemini 4 Hydrogen for everyday tasks; according to Google, it scores 51 on the Artificial Analysis Intelligence Index at the default high thinking level, surpassing the larger Argon's 50, and can reach 54 at the xhigh level. The model offers four thinking levels—low, medium, high, and xhigh—with 1 million tokens each of input and output context, priced at $1.50 per million input tokens and $7.50 per million output tokens, with cached hits at $0.075. Artificial Analysis's independent evaluation shows that Hydrogen at its highest level is roughly on par with Claude Sonnet 5.5, but lags behind Claude Sonnet 5.5 and GPT-6.1 Sol on terminal tasks such as Terminal-Bench 4.0.

  23. TechCrunch AISelected

    Google releases Gemini 4 Hydrogen model and upgrades Search

    Google has released the Gemini 4 Hydrogen model, calling it its biggest model improvement to Search, and the model has been rolled out to all Search users worldwide. Hydrogen targets fast-executing agentic workflows; the company says it completes tasks 2.5 times faster than Gemini 3.8 Flash and surpasses Gemini 4 Argon on the Artificial Analysis Intelligence Index at the default high thinking level. Search-side improvements include better understanding of multi-part questions and typos, optimized fan-out sub-query decomposition, and new visual answers in AI Mode such as images, comparison tables and custom tools; the model offers four thinking levels: low, medium, high and xhigh.

  24. Arena

    Arena Leaderboard: Gemini 4 Hydrogen Rises to No. 4 in Knowledge Work Tasks

    According to Arena.ai's statistics based on more than 6,000 real agent sessions, Gemini 4 Hydrogen (High) ranks No. 5 overall in Agent Arena and rises to No. 4 in knowledge work tasks. This ranking surpasses Claude Opus 5.5 (Max), GPT-6 Astra (Max), and Claude Sonnet 5.5 (Max), which rank higher on the overall leaderboard, and it is also the fastest model among the top five in Agent Arena.

  25. Unite.AISelected

    Google Releases Gemini 4 Hydrogen, Aimed at Everyday Agent Workflows

    On November 12, 2026, Google released Gemini 4 Hydrogen and began rolling it out globally across developer, enterprise, and consumer products, with paid API pricing at $1.50 per million input tokens and $7.50 per million output tokens. The model is derived from Gemini 4 Carbon and positioned for fast execution, covering coding, running tests, verifying changes, research, and tool-calling agents; it supports text, image, audio, and video input, with a knowledge cutoff of March 2026. The new Orchestrator option in Google Antigravity lets Carbon plan in a read-only phase while Hydrogen executes, and when it hits an obstacle it can hand back to Carbon to fix the plan.

  26. CursorFirst-partySelected

    Cursor launches Gemini 4 Hydrogen, $1.50/M input, $7.50/M output tokens

    Cursor announced that Gemini 4 Hydrogen is now available in Cursor, scoring 54.3% on CursorBench at default effort, higher than Gemini 3.8 Flash's 39.6%. The model is priced at $1.50/M input tokens and $7.50/M output tokens, and the official comparison link for each effort tier is provided.

  27. CursorFirst-partySelected

    Cursor launches Gemini 4 Hydrogen, scoring 54.3% on CursorBench

    Cursor announced that Gemini 4 Hydrogen is now available in Cursor, scoring 54.3% on CursorBench at the default effort, higher than Gemini 3.8 Flash's 39.6%. The model is priced at $1.50/M tokens for input and $7.50/M tokens for output, and the official team also provided a comparison entry point for each effort tier.

  28. Testing CatalogSelected

    Google releases Gemini 4 Hydrogen, 2.5x faster than 3.8 Flash

    Google has officially released Gemini 4 Hydrogen, which the company says completes tasks 2.5x faster than 3.8 Flash and features 4 thinking levels, including the newly added Extra level. Sundar Pichai said the model is aimed at everyday tasks, coding, and agentic workflows, and is rolling out today to Google AI Studio, the Gemini API, the Gemini App, Search, and paid Google AI Pro and Ultra subscribers, and is also launching on Google Antigravity. TestingCatalog says it surpasses Claude Sonnet 5.5 and GPT 6.1 Terra on most benchmarks.

  29. Demis HassabisSelected

    Google Releases Gemini 4 Hydrogen Model

    Google has released Gemini 4 Hydrogen, which the company says stands out in speed and intelligence, targeting everyday coding, agentic tasks, and general-purpose scenarios. The model is now available across all tiers of Google AI Studio, Gemini API, Gemini for business, the Gemini app, and Google Search, and paid Google AI Pro and Ultra subscribers can use it in Google Antigravity.

  30. Josh WoodwardSelected

    Google releases Gemini 4 Hydrogen and launches it on the Gemini App

    Google has released Gemini 4 Hydrogen, calling it frontier intelligence for everyone, and it is now available on the Gemini App. The publisher says the model is fast and smarter, marking a major leap forward.

  31. ArenaSelected

    Arena Evaluation: Gemini 4 Hydrogen Has the Fastest Step Latency Among the Agent Arena Top Ten

    Arena.ai states that Gemini 4 Hydrogen is the fastest model among the Agent Arena top ten, with a typical step latency of 7.3 seconds, about 2.7 times that of Claude Sonnet 5.5 and 3.1 times that of GPT-6.1 Sol. In Agent Arena sessions, 82% of Hydrogen's tasks were completed within 10 minutes, compared with 55% for GPT-6.1 Sol and 44% for Claude Sonnet 5.5; within 30 minutes, the completion rate was 97% for Hydrogen and 81% and 78% for the latter two. Arena says Hydrogen's overall improvement is comparable to these two models and that it did not trade quality for speed. Its measurement counts a single model API call within an Agent Mode task, timed from the first model request to the last streaming response, excluding tool execution.

  32. Artificial AnalysisSelected

    Google DeepMind releases Gemini 4 Hydrogen, adding an xhigh reasoning tier

    Google DeepMind releases Gemini 4 Hydrogen, adding an xhigh reasoning tier. It scores 54 on the Artificial Analysis Intelligence Index, 1 point lower than Claude Sonnet 5.5 (max) and GPT-6.1 Sol (max). Its per-task cost is about 32% of Sonnet 5.5's, but about 2.5 times that of GPT-6.1 Sol; the default high tier scores 51. The model supports a 1M token context window and accepts text, images, video, and speech as input, with pricing at $1.50/$7.50 per million input/output tokens and a 95% discount on cached input.

  33. Ammaar ReshiSelected

    Gemini 4 Hydrogen Released, Positioned as the Latest Flash Model

    Ammaar Reshi announced the release of Gemini 4 Hydrogen, calling it the latest Flash model that balances speed and intelligence and surpasses Sonnet 5.5 and 6.1 Terra in most respects. Starting that day, the model rolled out gradually on AI Studio, Antigravity, the Gemini app, and other entry points.

  34. Logan KilpatrickSelected

    Google releases the Gemini 4 Hydrogen model

    Google releases Gemini 4 Hydrogen, positioned as a highly capable and fast model, with the goal of bringing Gemini 4 to more users. The model is now available at all tiers of Google AI Studio, the Gemini API, Gemini for business, the Gemini app, and Google Search, and is open to Google AI Pro and Ultra paid subscribers in Google Antigravity.

  35. ArenaSelected

    Gemini 4 Hydrogen debuts at No. 5 on Agent Arena with a net improvement of 14.8%

    Google DeepMind's Gemini 4 Hydrogen (High) debuted on Agent Arena at No. 5, with a net improvement of 14.8% across 6.5k real agent sessions. The model is the fastest among the top 10, with a median time of 5.3 minutes per task—3 times faster than No. 1 Claude Fable 5.2 and 2.4 times faster than GPT-6 Astra; its median cost is $1.6 per task, 3.7 times cheaper than GPT-6 Astra and 4.7 times cheaper than Claude Fable 5.1, with API token pricing at $1.5 per million input tokens and $7.5 per million output tokens. By subcategory, it ranks No. 2 in controllability (+21%), No. 4 in positive feedback-to-complaint ratio (+32%), and No. 5 in confirmed success (+15%); by task category, it ranks No. 3 in Chat (+19%), No. 7 in Work (+13%), and No. 9 in Code (+12%).

  36. Google AISelected

    Google Releases Gemini 4 Hydrogen Model

    Google has added Gemini 4 Hydrogen to the Gemini 4 series, aimed at everyday coding and agentic workflows, with a speed 2.5 times that of Gemini 3.8 Flash. Google says that at the default thinking level it surpasses Argon on the Artificial Analysis Intelligence Index, and it offers adjustable thinking levels and can be used together with Gemini 4 Carbon in antigravity. The article also showcases the 3D Medic Managed Agent app built in GoogleAIStudio, which uses the model to inspect 3D files, find errors, and fix them before export.

  37. GeminiSelected

    Google releases the Gemini 4 Hydrogen model

    Google Gemini releases Gemini 4 Hydrogen, positioned as a fast everyday assistant for everyone. The company says the new model follows user instructions more strictly, can call the right tools, and delivers reliable output with fewer back-and-forth interactions.

  38. Philipp Schmid

    Google releases the Gemini 4 Hydrogen model, now available on AI Studio and the Gemini API

    Google has launched Gemini 4 Hydrogen on Google AI Studio and the Gemini API. The author says it has become the main model for daily fast execution and research loops, and is also suitable for use as a sub-agent paired with Carbon. The model scores 57% on Terminal-Bench 4.0, surpasses Argon on Artificial Analysis, offers four thinking levels—low, medium, high, and xhigh—and is priced at $1.50 per million input tokens and $7.50 per million output tokens, with free quotas available on Google AI Studio and the Gemini API. It is also available to paid Google AI Pro and Ultra subscribers, covering antigravity, the Gemini App, Google Search, and Google Cloud.

  39. Sundar PichaiSelected

    Google releases Gemini 4 Hydrogen, completing tasks 2.5 times faster than 3.8 Flash

    Sundar Pichai announced the release of Gemini 4 Hydrogen, aimed at everyday tasks, coding, and agentic workflows, completing tasks 2.5 times faster than 3.8 Flash. The model offers 4 levels of thinking depth, letting users pay and wait only for the reasoning depth the task requires. Starting that day, it is available across all tiers of Google AI Studio, the Gemini API, the Gemini App, Search, and Google Antigravity to paid Google AI Pro and Ultra subscribers.

  40. 9to5GoogleSelected

    Google releases Gemini 4 Hydrogen, claiming major improvements in both speed and intelligence

    Google has released Gemini 4 Hydrogen, replacing the Flash brand, and describes it as a standalone major-version upgrade over Gemini 3.8 Flash, supporting native 1 million token input and up to 1 million token output. The company claims it completes tasks 2.5 times faster under high default settings, scores 57% on Terminal-Bench 4.1 (compared with 18.2% for 3.8 Flash), and reduces fabricated statements by more than 5 times. The model has entered the Gemini app model selector, become the default model for Google Search AI Mode and Antigravity, and is priced for developers at $1.50 per million input tokens and $7.50 per million output tokens.

  41. Testing Catalog

    Gemini 4 Hydrogen (Flash) launches on Antigravity

    Gemini 4 Hydrogen (Flash) is rolling out on Antigravity, and users need to update to the latest version to try it. Cited information states that the model is also available to Pro subscribers.

Nov 12

  1. Hacker News TrendingSelected

    Google Antigravity Launches Gemini 4 Hydrogen and Orchestrator Scheduling Mode

    Google Antigravity has launched Gemini 4 Hydrogen and added an Orchestrator option to the model selector, which combines and schedules Hydrogen with Gemini 4 Carbon. Orchestrator dynamically switches between the two models within a single task trajectory: Carbon handles exploration and planning, Hydrogen handles code changes and test runs, and when obstacles are encountered, control is handed back to Carbon to fix the plan; on Terminal-Bench 4.1 it achieves a task success rate close to Carbon while using an average of 24% less quota and being 19% faster. Hydrogen can also be used on its own, scoring 51 on the Artificial Analysis Intelligence Index and 57% on Terminal-Bench 4.1, and it offers four reasoning depth levels: low, medium, high, and xhigh.

  2. Google DeepMindFirst-partySelected

    Google DeepMind releases Gemini 4 Hydrogen, targeting everyday coding and agentic workflows

    Google DeepMind has released Gemini 4 Hydrogen, focused on speed and efficiency for everyday coding and agentic workflows. On the Artificial Analysis Intelligence Index, it scores 51 at the default high thinking level, above Argon's 50, and can reach 54 at the xhigh level. The model's 1 million token context window is priced uniformly at $1.50 per million input tokens and $7.50 per million output tokens, and it offers four thinking levels: low, medium, high, and xhigh. Google has also launched Orchestrator in Antigravity, which can dynamically switch models between Carbon and Hydrogen within the same session; on Terminal-Bench 4.0 it approaches Carbon's task success rate while consuming an average of 24% less quota and running 19% faster.

  3. Google BlogFirst-partySelected

    Google DeepMind releases Gemini 4 Hydrogen, targeting everyday coding and agentic workflows

    Google DeepMind has released Gemini 4 Hydrogen, focused on speed and efficiency for everyday coding and agentic workflows. It scores 51 on the Artificial Analysis Intelligence Index at the default high thinking level, above Argon's 50, and reaches 54 at the highest xhigh level. The model has unified pricing across its 1 million token context window, at $1.50 per million input tokens and $7.50 per million output tokens, and offers four thinking levels: low, medium, high, and xhigh. Google Antigravity has simultaneously launched Orchestrator, which can dynamically switch between Carbon and Hydrogen within the same session, achieving task success rates on Terminal-Bench 4.0 close to Carbon while consuming an average of 24% less quota and running 19% faster. Hydrogen is available starting today through Gemini API, Google Antigravity, the Gemini app, and other entry points for developers, enterprises, and consumers.

  4. Google BlogFirst-partySelected

    Google releases Gemini 4 Hydrogen, aimed at coding and agentic workflows

    Google releases Gemini 4 Hydrogen, aimed at everyday coding and agentic workflows, supporting repository-level coding and debugging, interactive web development, and multimodal workflows. The model can run standalone and supports four configurable thinking levels, and can also be used in combination with frontier models such as Gemini 4 Carbon and Gemini 4 Argon; the company says it follows instructions more closely and fabricates less content.

  5. Google BlogFirst-partySelected

    Google Search Integrates Gemini 4 Hydrogen and Opens It to Users Worldwide

    Google has integrated Gemini 4 Hydrogen into Google Search and opened it to all users worldwide; the company calls this the largest model upgrade to Search to date. After the update, Search can understand multi-part complex questions, spelling errors, and highly specific requests, and responds to each point through stronger instruction following and tool calling. Hydrogen's reasoning capabilities also enhance query expansion, retrieving multiple related subtopics in parallel, returning more images, structured answers, comparison tables, and custom tools, and covering all supported AI Mode languages.

Heat trend for this story

Current
12
Period peak
282 Nov 13 10:00
24-hour change
-92%

Move the pointer or select the chart to inspect hourly heat. Use the left and right arrow keys with a keyboard.

Related stories

  1. Related storiesMeta launches Muse Wonder and Muse Wonder Full models
  2. Related storiesAnthropic releases Claude Opus 5.5
  3. Related storiesEnglish translation is being prepared.
  4. Related storiesEnglish translation is being prepared.