The company is positioning Inkling as a flexible base model for customization. “Inkling is not the strongest overall model available today,” the announcement<\/a>\u00a0states. Thinking Machines expects the mix of multimodal support, efficient processing, and fine-tuning options to set the model apart.<\/p>\n Thinking Machines says it pre-trained Inkling on 45 trillion tokens of public and synthetic text, images, audio recordings, and videos. The training set also\u00a0includes public data that “may\u00a0be subject to intellectual property protection.”<\/a> The company used the Chinese AI model Kimi K2.5, among other methods, to generate synthetic data. Kimi K2.5 also served as the basis for Cursor’s coding model<\/a>. More technical details are available in\u00a0the model card<\/a>.<\/p>\n Inkling leads U.S. open models but trails China’s best<\/p>\n According to AI benchmarking platform Artificial Analysis<\/a>, Inkling debuts with a score of 41 on the Artificial Analysis Intelligence Index. That makes it the leading open-weights model from a U.S. lab. It ranks three points above the previous leader, Nemotron 3 Ultra at 38, and well ahead of Gemma 4 31B at 29 and gpt-oss-120b at 24.<\/p>\n On GDPval-AA v2, an agent-based benchmark that simulates knowledge-work tasks, Inkling reaches an Elo rating of 1,238. It beats Kimi K2.6 at 1,190 and DeepSeek v4 Flash max at 1,189. Inkling also scores 24 percent on the Tau-3 banking benchmark, ahead of Kimi K2.6 at 21 percent and DeepSeek v4 Flash max at 23 percent.<\/p>\n Inkling performs rather poorly on factual accuracy. Artificial Analysis gives the model a score of just +2 on its AA Omniscience benchmark. That puts it below the leading open-weights models, though still above other U.S. models such as Nemotron 3 Ultra at -1. Inkling’s accuracy is 40 percent, while its hallucination rate is 63 percent. Those results are likely to limit its use in applications that need highly accurate information.<\/p>\n With a 64K context window, Inkling costs $1.87 per million input tokens and $4.68 per million output tokens. That’s slightly more than open-source Chinese models such as GLM-5.2 and DeepSeek v4, which offer similar or better performance on text and code tasks. For context windows up to 256,000 tokens, pricing rises to $3.74 for input, $0.748 for cached input, and $9.36 for output.<\/p>\n But Inkling uses fewer output tokens than comparable open-weights models. According to Artificial Analysis, it averages 25,000 output tokens per Intelligence Index task. GLM-5.2 max uses 43,000, Kimi K2.6 uses about 38,000, and DeepSeek v4 Pro max uses about 37,000 tokens on the same tasks.<\/p>\n Thinking Machines says Inkling offers continuously adjustable “thinking effort.” Users can choose their preferred balance between cost and performance, reducing token use while maintaining the same result quality.<\/p>\n Inkling-Small beats the larger model on some benchmarks<\/p>\n![\"Inkling](\"https:\/\/www.europesays.com\/ai\/wp-content\/uploads\/2026\/07\/inkling_AA_overall_performance-scaled-1.jpg\")
Inkling scores 41 on the Artificial Analysis Intelligence Index, making it the leading U.S. open-weights model. | Image: Artificial Analysis<\/p>\n![\"\"](\"https:\/\/www.europesays.com\/ai\/wp-content\/uploads\/2026\/07\/inkling_AA_GDPval_overall_performance-scaled-1.jpg\")
Inkling outperforms Kimi K2.6 and DeepSeek v4 Flash max on agent-based knowledge-work tasks. | Image: Artificial Analysis<\/p>\n![\"\"](\"https:\/\/www.europesays.com\/ai\/wp-content\/uploads\/2026\/07\/inkling_AA_GDPval_token_performance-scaled-1.jpg\")
Inkling scores +2 on AA Omniscience, with 40 percent accuracy and a 63 percent hallucination rate. | Image: Artificial Analysis<\/p>\n