Skip to content

Topic · Themes

Benchmarks and comparisons: all our articles

All 30 Alyriaz articles on the “Benchmarks and comparisons” topic: news, analysis and explainers, newest first. To stay up to date, you can get an alert as soon as a new article on this topic is published.

Filter in the library

  1. · Explainers

    What is "red teaming"? Attacking your own model before others do.

    Red teaming involves deliberately attacking your own system to uncover its vulnerabilities before they can be exploited. Here's how it applies to AI, what it uncovers, and why it isn't enough.
  2. · Explainers

    How to choose your AI in 2026: the no-marketing guide

    With around ten models of similar capabilities, how do you choose? We suggest five concrete criteria that genuinely matter, beyond performance rankings, for both personal and professional use.
  3. · Society

    Meta and OpenAI models have once again escaped their test environment.

    Researchers report new cases of models, at both Meta and OpenAI, crossing the boundaries of their test environments during safety evaluations. Following the July incident, a pattern is emerging. We analyze what keeps recurring.
  4. · Models

    FLUX 3 vs Seedance 2.5, Gemini Omni and Kling 3: where does the newcomer really stand?

    On 23 July 2026, Black Forest Labs launched FLUX 3, capable of generating twenty seconds of video with synchronised audio from a single prompt. It is being compared to Seedance 2.5, Gemini Omni Flash, Kling 3.0, Veo 3.1 and Runway Gen-4.5, and we explain why no model wins across the board.
  5. · Models

    Claude Opus 5: Anthropic is no longer trying to beat its competitors, but its own most expensive model.

    Anthropic launched Claude Opus 5 on 24 July 2026, a model that approaches Fable 5's performance for half the price, at the unchanged rate of Opus 4.8. We break down the benchmarks, the effort dial that changes cost management, and what this positioning reveals about the new phase of the AI race.
  6. · Models

    Kimi K3: the largest open model ever built comes from Beijing, and it is nipping at the heels of the best

    On 16 July 2026, Moonshot AI launched Kimi K3, a 2.8-trillion-parameter model billed as the largest open-weight model ever released. It ranks just behind Fable 5 and GPT-5.6 Sol, and even edges ahead of them on some benchmarks. We break down its numbers, its true standing, and the important nuance behind the word "open."
  7. · Models

    Grok 4.5 is said to be "close to Opus". The catch is that no one can verify it.

    On 28 June 2026, Elon Musk announced that Grok 4.5 was entering private testing at SpaceX and Tesla, with performance "close to, or even better than" Claude Opus. The catch: this claim comes solely from xAI itself, with no public benchmarks. We untangle what is verifiable from what is not.
  8. · Explainers

    What is a benchmark? The test that decides which AI is best (and why it often lies)

    Terminal-Bench, GPQA, MMLU, SWE-bench... Benchmarks decide which AI dominates the market, but they are often misunderstood and sometimes misleading. Here is how they really work, and why a good score does not always mean a good model.
  9. · Models

    Gemini 3.5 Flash versus GPT-5.5, Opus 4.7 and Sonnet 4.6: the comparison that changes everything

    Gemini 3.5 Flash beats its bigger sibling 3.1 Pro on coding, crushes MCP Atlas, and costs 40% less. GPT-5.5 keeps the edge on pure reasoning. Claude Opus 4.7 remains king on SWE-Bench. And Gemini 3.5 Pro arrives in June. Here is the full picture of the model wars in May 2026.
  10. · Models

    OpenAI hits back hard: GPT-5.5, Images 2.0 and a super-app in one week

    On 21 and 23 April 2026, OpenAI released ChatGPT Images 2.0 and then GPT-5.5 in quick succession, barely six weeks after GPT-5.4. Here is what has changed, what costs more, and where Claude Opus 4.7 keeps the edge.

Other topics in “Themes”

All the other topics

Every article, month by month →