Artificial Analysis now displays base model logos alongside derived video models on its leaderboards and has added an option to hide derived models to rank base models only.
Vidu has released Vidu Q4 Preview ahead of its full Q4 release, generating 3 to 16 second clips up to 4K with native audio and placing third on the Artificial Analysis AA-Video-I2V v1.0 leaderboard.
Artificial Analysis has launched AA-Music-Vocal v1.1 and AA-Music-Instrumental v1.1 for its Music Arena, evaluating models across 1,000 new prompts in 17 genres. Suno v6 ranks first on both leaderboards, followed by Suno v6-mini in second.
Artificial Analysis scored Claude Haiku 5.5 at 43, up 26 points from the previous Haiku, and reported a 1 million token context window. Its results also cover token use, terminal tasks, factual accuracy, and an AutomationBench score the firm says is likely understated.
Google's Gemini 3.8 Live Extended Thinking model has taken first place on the Artificial Analysis Speech to Speech Index, while the standard version placed second in blind live human evaluations. The standard model costs $0.84 per hour of input audio, the lowest rate listed on the index.
Xiaomi addressed an issue where models trained only to pass tests added unrequested code and let errors pass silently by multiplying test results by quality checklist scores. The resulting MiMo-V2.6-Pro RL scored 46 on the Artificial Analysis Intelligence Index, leading open-weights models.
Artificial Analysis evaluated Mistral Large 4 Preview, which scored 38 on the Intelligence Index and 50% on the Cyber Index. Mistral AI is offering the 512k-context model at $1.36 per 1M input tokens and $4.18 per 1M output tokens, with a 50% discount during its first two weeks.
Ant Group has released Ling 3.1 Flash, a reasoning model featuring 560 billion total parameters and a 1 million token context window. Artificial Analysis evaluated the model at 41 on its Intelligence Index v4.3, highlighting notable improvements in agentic performance and token efficiency.
Artificial Analysis benchmarked OpenAI Web Search at roughly 31 seconds of Time per Task, ranking 14th among 26 tested products. Because OpenAI does not report search latency directly, the team measured it through stream events.
Artificial Analysis benchmarked OpenAI Web Search on its Search Index, where the tool scored 74 to place as the 5th best provider and 7th among 26 tested products. Running on GPT-5.6 Luna with medium reasoning, the built-in search tool averaged about $0.05 per task.
Artificial Analysis reports that ElevenLabs' Eleven v4 Turbo processes 96 characters per second of generation time at $40 per 1 million characters. The benchmark shows the model runs faster than Eleven v4 while costing half the price.
Artificial Analysis ranks ElevenLabs’ Eleven v4 Turbo first on its Provider Voice TTS Arena, with an Elo of 1,334 ahead of Eleven v4, Qwen-Audio-3.1-TTS-Plus, and Sonic 3.6. It places second on Controlled Voice and scores 90.1% on pronunciation robustness.
Artificial Analysis has open-sourced AA-AgentPerf-Local, which replays recorded agent sessions to measure local inference speed, and published initial 4-bit results for four systems. On models that fit in 32 GB, the GeForce RTX 5090 finished more than 3.5 times faster than the other systems tested.
Artificial Analysis says that one year ago GPT-5 mini (high) led its Intelligence Index at 17 on v4.3 and about $0.05 per task. As of last week, Claude Opus 5.5 holds the highest score to date at 58 and about $5.98 per task.
Artificial Analysis says GPT-6.1 Sol replaces GPT-6 Sol after seven days, gains 4 Intelligence Index points on GPT-6 Sol, and costs $0.72 per task at max effort versus $3.26 for GPT-6 Astra. Its list price matches GPT-6 Sol at $2/$10 per million tokens, with the cache-read discount rising from 90% to 95%.
Artificial Analysis says Claude Sonnet 5.5 scores 56 on its Intelligence Index at max effort, two points behind Claude Opus 5.5 (max), while using the highest output-token count it has measured and costing about $7.60 per task.
Artificial Analysis says ElevenLabs’ Eleven v4 text-to-speech model, with support for more than 90 languages, ranks first on its Provider Voice arena and Pronunciation Robustness benchmark, and second on Controlled Voice.
Artificial Analysis says the Cyber Index is the average of CWE-Bench-AA, DeepsecBench-AA, and CyberGym-E2E-AA. Tasks a model declines on safety grounds score zero and are reported separately from other failures.
Artificial Analysis launched the Cyber Index and Cyber Index Alliance with CollinearAI, IBM, NVIDIA, and Vercel. The index combines three open benchmarks that score how AI agents find and fix vulnerabilities, with Grok 4.7 (xhigh) and MiMo-V2.6-Pro leading at 56.
Artificial Analysis says Claude Opus 5.5 uses 15.6 million tokens per Coding Agent Index task, compared with 11.4 million for Opus 5. Cached input rises from 10.9 million to 14.6 million, and output more than doubles from 137,000 to 333,000.