xAI · Models · Pricing and access
Grok Voice Transcribe 2.0 ranks first for accuracy among 32 models
This story was updated after publication (Oct. 9, 2026). Details
xAI released it on September 18. It sits first for accuracy among 32 streaming models on the independent Artificial Analysis leaderboard. By the company’s own measurement, word error rate on short phrases fell from 20.6% to 6.8% and on conversational audio from 8.7% to 3.3%. Pricing is unchanged from 1.0: $0.10 per hour of audio for batch, $0.20 for streaming. The model handles dozens of languages and follows a mid-recording language switch in a single pass.
Sources
- xAI (SpaceXAI), “Introducing Grok Voice Transcribe 2.0”, (x.ai)
- Artificial Analysis, “Speech to Text Providers Leaderboard & Comparison”, streaming speech-to-text leaderboard (artificialanalysis.ai)
Corrections and updates
- The Turkish version first posted on Instagram described the $0.10-per-hour price as being for batch translation. The model does transcription, not translation; the price is for batch transcription, and the Turkish text has been corrected. The English text was accurate.
- After this story was published, Microsoft AI’s MAI-Transcribe-2-Streaming, released on October 1, took first place on the Artificial Analysis streaming leaderboard. As of October 9, Grok Voice Transcribe 2.0 ranks second, and the leaderboard has 45 entries.
About this story
This story was posted on Instagram by @jarrus.tech on Sept. 23, 2026.
Spotted an error in this story? [email protected] · Instagram
Short link: thejarrus.com/en/h7-41
This story in Turkish: Grok Voice Transcribe 2.0, doğrulukta 32 model arasında birinci
On the same topic
Anthropic releases Claude Haiku 5.5 at 90% lower prices
Claude Haiku 5.5, released Oct. 7, costs 90% less than Haiku 4.5 for prompts up to 100,000 tokens. It leads on benchmarks but uses more tokens per task.
OpenAI is rolling out GPT-6 to everyone in ChatGPT
OpenAI began rolling out GPT-6 to everyone in ChatGPT on October 7. Its new Intelligent UI builds answers from text, visuals and interactive elements.
The open model war between Europe, the US and China heats up
Mistral unveiled Large 4 on October 6. The same week, Ecosia dropped Mistral, Reflection AI launched Beam and DeepSeek was reported near a $12 billion round.