What changed in AI
Every material change the connectors observe becomes an event with a source. Dates are when the change occurred (effective date when known); hover a date for the observation time. Older events load with a cursor.
1,145 events since 4 Jun 2026 · ordered by occurrence
15 Sept 2026 3
Qwen3 Next 80B A3B Thinking: max output tokens changed from 235929 to 32768
Max output235.9K tokens→32.8K tokensopenrouterQwen3.5-35B-A3B: max output tokens changed from 16384 to 65536
Max output16.4K tokens→65.5K tokensopenrouterZ.ai GLM 5.2: max output tokens changed from 131072 to 128000
Max output131.1K tokens→128K tokensopenrouter
14 Sept 2026 44
pplx-pii-masking-vllm: release date changed from 2026-09-13 to 2026-08-02
Release date13 Sept 2026→2 Aug 2026huggingfacepplx-pii-masking-vllm: license changed from MIT to Other
LicenseMIT→Otherhuggingfacepplx-pii-masking-vllm: license raw changed from mit to other
License rawmit→otherhuggingfacepplx-pii-masking-vllm: last modified changed from 2026-09-13T22:30:27+00:00 to 2026-08-02T23:54:58+00:00
Last modified2026-09-13T22:30:27+00:00→2026-08-02T23:54:58+00:00huggingfaceHermes 3 70B Instruct: supported parameters changed from ["frequency_penalty", "logit_bias", "max_tokens", "min_p"… to ["frequency_penalty", "logit_bias", "max_tokens", "min_p"…
Supported parametersfrequency_penalty, logit_bias, max_tokens, min_p, presence_penalty, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, top_k, top_p→frequency_penalty, logit_bias, max_tokens, min_p, presence_penalty, repetition_penalty, seed, stop, structured_outputs, temperature, top_k, top_popenrouterLlama 3.1 Euryale 70B v2.2: supported parameters changed from ["frequency_penalty", "logit_bias", "max_tokens", "min_p"… to ["frequency_penalty", "logit_bias", "max_tokens", "min_p"…
Supported parametersfrequency_penalty, logit_bias, max_tokens, min_p, presence_penalty, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_p→frequency_penalty, logit_bias, max_tokens, min_p, presence_penalty, repetition_penalty, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_popenrouterLlama Guard 4 12B: structured output changed from true to false
Structured outputYes→NoopenrouterLlama Guard 4 12B: supported parameters changed from ["frequency_penalty", "logit_bias", "max_tokens", "min_p"… to ["frequency_penalty", "logit_bias", "max_tokens", "min_p"…
Supported parametersfrequency_penalty, logit_bias, max_tokens, min_p, presence_penalty, repetition_penalty, response_format, seed, stop, temperature, top_k, top_p→frequency_penalty, logit_bias, max_tokens, min_p, presence_penalty, repetition_penalty, seed, stop, temperature, top_k, top_popenrouterNemotron 3.5 Content Safety: structured output changed from true to false
Structured outputYes→NoopenrouterNemotron 3.5 Content Safety: supported parameters changed from ["frequency_penalty", "include_reasoning", "logit_bias", … to ["frequency_penalty", "include_reasoning", "logit_bias", …
Supported parametersfrequency_penalty, include_reasoning, logit_bias, max_tokens, min_p, presence_penalty, reasoning, repetition_penalty, response_format, seed, stop, temperature, top_k, top_p→frequency_penalty, include_reasoning, logit_bias, max_tokens, min_p, presence_penalty, reasoning, repetition_penalty, seed, stop, temperature, top_k, top_popenrouterInkling: structured output changed from true to false
Structured outputYes→NoopenrouterInkling: supported parameters changed from ["frequency_penalty", "include_reasoning", "logit_bias", … to ["frequency_penalty", "include_reasoning", "logit_bias", …
Supported parametersfrequency_penalty, include_reasoning, logit_bias, max_tokens, min_p, presence_penalty, reasoning, reasoning_effort, repetition_penalty, response_format, seed, stop, temperature, tool_choice, tools, top_k, top_p→frequency_penalty, include_reasoning, logit_bias, max_tokens, min_p, presence_penalty, reasoning, reasoning_effort, repetition_penalty, seed, stop, temperature, tool_choice, tools, top_k, top_popenrouterLing 3.0 Flash Fin: supported parameters changed from ["frequency_penalty", "include_reasoning", "logit_bias", … to ["frequency_penalty", "include_reasoning", "logit_bias", …
Supported parametersfrequency_penalty, include_reasoning, logit_bias, max_tokens, min_p, presence_penalty, reasoning, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_p→frequency_penalty, include_reasoning, logit_bias, max_tokens, min_p, presence_penalty, reasoning, repetition_penalty, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_popenrouterNew model: K2 Horizon 7B (MBZUAI Institute of Foundation Models)
artificial_analysisNew model: K2 Horizon 3.7B (MBZUAI Institute of Foundation Models)
artificial_analysisNew model: K2 Horizon MoVA 36B A4B (MBZUAI Institute of Foundation Models)
artificial_analysisNew model: K2 Horizon 0.9B (MBZUAI Institute of Foundation Models)
artificial_analysisLlama 4 Maverick: max output tokens changed from 115200 to 16384
Max output115.2K tokens→16.4K tokensopenrouterCohereLabs/command-a-plus-05-2026-fp8: base model changed from ["CohereLabs/command-a-plus-05-2026"] to ["CohereLabs/command-a-plus-05-2026", "CohereLabs/command…
Base modelCohereLabs/command-a-plus-05-2026→CohereLabs/command-a-plus-05-2026, CohereLabs/command-a-plus-05-2026-bf16huggingfacemlx-community/Qwen3.8-27B-4bit: last modified changed from 2026-08-14T15:52:00+00:00 to 2026-09-14T12:37:43+00:00
Last modified2026-08-14T15:52:00+00:00→2026-09-14T12:37:43+00:00huggingfaceNemotron 3 Super: max output tokens changed from 16384 to 235929
Max output16.4K tokens→235.9K tokensopenrouterCohereLabs/command-a-plus-05-2026-fp8: base model changed from ["CohereLabs/command-a-plus-05-2026", "CohereLabs/command… to ["CohereLabs/command-a-plus-05-2026"]
Base modelCohereLabs/command-a-plus-05-2026, CohereLabs/command-a-plus-05-2026-bf16→CohereLabs/command-a-plus-05-2026huggingfaceLing 3.0 Flash VL: hf inference providers changed from ["novita"] to ["deepinfra", "novita"]
Hf inference providersnovita→deepinfra, novitahuggingfaceNew artifact: XiaomiMiMo/MiMo-VL-7B-RL-GGUF (Xiaomi)
huggingfaceNew artifact: CohereLabs/command-a-plus-05-2026-fp8 (Cohere)
huggingfaceNew model: FLUX.1-Redux-dev (Black Forest Labs)
huggingfaceQwen3-VL-4B-Instruct: reasoning changed from true to false
ReasoningYes→Noartificial_analysispplx-pii-masking: last modified changed from 2026-09-02T17:08:49+00:00 to 2026-09-14T02:30:28+00:00
Last modified2026-09-02T17:08:49+00:00→2026-09-14T02:30:28+00:00huggingfacegpt-5: reasoning changed from false to true
ReasoningNo→Yesopenrouterperplexity-ai/pplx-pii-masking-GGUF: tags changed from ["llama.cpp", "gguf", "pii", "token-classification", "fea… to ["en", "feature-extraction", "gguf", "llama.cpp", "multil…
Tagsllama.cpp, gguf, pii, token-classification, feature-extraction→en, feature-extraction, gguf, llama.cpp, multilingual, pii, privacy, qwen3, token-classificationhuggingfaceperplexity-ai/pplx-pii-masking-GGUF: license changed from Other to MIT
LicenseOther→MIThuggingfaceperplexity-ai/pplx-pii-masking-GGUF: license raw changed from other to mit
License rawother→mithuggingfaceperplexity-ai/pplx-pii-masking-MLX: tags changed from ["mlx", "pii", "token-classification", "privacy", "apple-… to ["apple-silicon", "en", "mlx", "multilingual", "pii", "pr…
Tagsmlx, pii, token-classification, privacy, apple-silicon→apple-silicon, en, mlx, multilingual, pii, privacy, qwen3, token-classificationhuggingfaceperplexity-ai/pplx-pii-masking-MLX: license changed from Other to MIT
LicenseOther→MIThuggingfaceperplexity-ai/pplx-pii-masking-MLX: license raw changed from other to mit
License rawother→mithuggingfaceNew model: pplx-pii-masking-vllm (Perplexity AI)
huggingfaceperplexity-ai/pplx-pii-masking-GGUF: last modified changed from 2026-07-16T23:51:56+00:00 to 2026-09-13T22:44:08+00:00
Last modified2026-07-16T23:51:56+00:00→2026-09-13T22:44:08+00:00huggingfaceperplexity-ai/pplx-pii-masking-MLX: last modified changed from 2026-07-17T00:03:56+00:00 to 2026-09-13T22:44:09+00:00
Last modified2026-07-17T00:03:56+00:00→2026-09-13T22:44:09+00:00huggingfaceQwen3 14B: max output tokens changed from 8192 to 16384
Max output8.19K tokens→16.4K tokensopenrouterNemotron 3 Ultra: max output tokens changed from 32768 to 182520
Max output32.8K tokens→182.5K tokensopenroutergpt-5.6-luna: reasoning changed from false to true
ReasoningNo→YesopenrouterGLM 5.3: max output tokens changed from 131072 to 943717
Max output131.1K tokens→943.7K tokensopenrouter
13 Sept 2026 3
Z.ai GLM 5.2: max output tokens changed from 182476 to 131072
Max output182.5K tokens→131.1K tokensopenrouterQwen3 VL 235B A22B Instruct: reasoning changed from true to false
ReasoningYes→Noartificial_analysis
Feed keyed on occurred_at. Backfill = history imported when a source is first crawled (excluded — tick the box in the rail to include it). Times are UTC.