Skip to content
A Spellwright Studios publication
Spellwright News

AI Intelligence

What's happening in AI, in one place

The companies building AI models, every model we track from newest to oldest, what's coming next, and how they score on the benchmarks people use to compare them. Open-weight and closed models side by side, with a source for every fact.

Checked through Nov 1, 2026.

Companies
9
Models
39
Open-weight models
8
Benchmark results
42

Newest models

Full timeline of 39 models

Upcoming

  • Gemini 4 CarbonGoogle · announced Oct 28, 2026Upcoming

    Google announced Gemini 4 Carbon on October 28, 2026, with a phased rollout to selected cyber defenders and trusted partners. Wider access is forthcoming; a public API model identifier and general-release date are not established.

Benchmarks

Scores as published. means the model's maker reported it. Makers test under different settings, so treat close scores as close. Dark bars are closed models; light bars are open-weight models.

Terminal-Bench 4.0

Coding · higher is better · 5 models
  1. Claude Sonnet 5.570.6%*
  2. Claude Opus 5.566.4%*
  3. Claude Opus 552.3%*
  4. Gemini 3.8 Flash19.1%*
  5. Claude Sonnet 510.3%*

CursorBench 4.0

Coding · higher is better · 4 models
  1. Claude Opus 5.557.8%*
  2. Claude Sonnet 5.555.5%*
  3. Claude Opus 546.6%*
  4. Claude Sonnet 534.1%*

Humanity's Last Exam

Reasoning and knowledge · higher is better · 4 models
  1. Claude Opus 5.567.7%*
  2. Claude Sonnet 5.564.5%*
  3. Claude Opus 563.6%*
  4. Claude Sonnet 554.9%*

HellaSwag

Reasoning and knowledge · higher is better · 4 models
  1. GPT-4 pretrained base (paper model)95.3%*
  2. PaLM 540B (paper model)83.8%*
  3. Mistral-7B-v0.181.3%*
  4. Llama 2 7B pretrained (July 2023 release)77.1%*

All 5 results

All 18 benchmarks

Scoreboard

Best published score per model, newest models first. A dash means no score has been published that we have checked.

ModelTerminal-Bench 4.0CursorBench 4.0Humanity's Last ExamHellaSwagHumanEvalDeepSWE 1.1
Gemini 4 Argon –––––77.9%
Claude Sonnet 5.5 70.6%55.5%64.5%–––
Claude Opus 5.5 66.4%57.8%67.7%–––
Claude Opus 5 52.3%46.6%63.6%–––
Claude Sonnet 5 10.3%34.1%54.9%–––
Mistral-7B-v0.1 · open–––81.3%30.5%–
Llama 2 7B pretrained (July 2023 release) · open–––77.1%––
GPT-4 pretrained base (paper model) –––95.3%67.0%–
PaLM 540B (paper model) –––83.8%76.2%–
Gemini 3.8 Flash 19.1%––––73.7%

Companies

  • Google

    11 models · newest: Gemini 4 Carbon

    Google provides the Gemini family of AI models, including the Gemini 4 Argon release and the announced Gemini 4 Carbon rollout.

  • Anthropic

    11 models · newest: Claude Sonnet 5.5

    Anthropic provides the Claude family of AI models, including Claude Sonnet 5.5.

  • Qwen Team

    2 models · newest: Qwen2.5-7B

    Qwen Team develops the Qwen series of models, including the Qwen2.5 language models. Its model repository licences name Alibaba Cloud in copyright notices. The inspected pages do not establish a legal operator or corporate parent mapping.

  • OpenAI

    9 models · newest: GPT-4o mini (July 2024 release)

    OpenAI provides the API models documented here, including GPT-6 Luna.

  • Mistral AI

    2 models · newest: Mistral-7B-Instruct-v0.1

    Mistral AI develops language models, including the Mistral 7B base model and its instruction-tuned variant. Its current website legal notice identifies a French company named Mistral. Share-listing status, ultimate ownership and financial details are not established in this record.

  • Meta

    2 models · newest: Llama 2 7B Chat (July 2023 release)

    Meta developed Llama 2, a family of pretrained and dialogue-tuned language models released under a custom community licence. This provider identity entry does not establish its corporate ownership chain or finances.

  • Alphabet

    0 models

    Alphabet Inc. is the Nasdaq-listed issuer reporting consolidated Google Services, Google Cloud and other businesses in this February 2026 earnings release. Its consolidated figures are not a measure of Gemini-only revenue or AI-only spending. Legal jurisdiction and a detailed subsidiary ownership chain are unassessed here.

  • BigScience Workshop

    1 model · newest: BLOOM 176B

    BigScience Workshop is an international research collaboration credited with developing BLOOM. Its project licence describes a collaborative open-innovation effort; the paper describes hundreds of researchers. This record does not identify it as a subsidiary or establish a legal corporate entity, jurisdiction or financial accounts.

  • EleutherAI

    1 model · newest: GPT-NeoX-20B

    EleutherAI developed GPT-NeoX-20B, a research-oriented autoregressive language model trained on the Pile.

How this is built

Sol, our AI editor, collects each fact from the company's own announcement, model card, price page or filing, or from the benchmark's authors, and links it. "Open weights" means the company released the model weights for download; it doesn't by itself mean an open-source license. Release dates are the first day the public could use a model. Upcoming models are announced but not yet available. Missing data is shown as missing, never as zero.