Explore Franklin AI

GPT-6 Astra (Medium)

Benchmark scores, pricing, speed and comparisons for GPT-6 Astra (Medium).

Key scores

Intelligence49.6Index score
Coding76.7Index score
Speed49.7Tokens/sec

GPT-6 Astra (medium) is an artificial intelligence model developed by OpenAI and released on September 3, 2026. In terms of performance metrics, the model holds an intelligence index of 59.2 and a coding index of 76.7, while its math index remains unknown. Benchmark results for the model include a GPQA score of 0.939, an HLE score of 0.527, a SciCode score of 0.51, and an LCR score of 0.763333333333333. Regarding pricing, the model is available at a rate of $10.00 per million tokens for input and $50.00 per million tokens for output, resulting in a blended cost of $20.00 per million tokens. Specific data regarding output speed and time to first token are currently unavailable.

No benchmark data available for this model.

Comparisons with GPT-6 Astra (Medium)

Generated Franklin AI comparison pages for this model.

Model ComparisonGPT-6 Astra vs. Claude Fable 5.1: A Comparative AnalysisReleased within days of each other in September 2026, OpenAI’s GPT-6 Astra and Anthropic’s Claude Fable 5.1 represent the current frontier of AI capability. While both models share identical pricing structures, they diveModel ComparisonGPT-6 Astra vs. Grok 4.6: A Comparative AnalysisThis analysis evaluates the performance and economic tradeoffs between OpenAI’s GPT-6 Astra and SpaceXAI’s Grok 4.6. While Astra leads in raw intelligence and coding benchmarks, Grok 4.6 offers significant cost advantageModel ComparisonClaude Haiku 5.5 (Max) vs GPT-6 Astra (Medium)Claude Haiku 5.5 Max has lower listed token prices than GPT-6 Astra Medium. Astra Medium has the higher intelligence index and a much shorter reported first-token delay, while Haiku has higher reported output throughput.Model ComparisonGrok 4.7 (Low) vs GPT-6 Astra (Medium)Grok 4.7 Low has lower listed token rates than GPT-6 Astra Medium, a shorter reported first-token delay and higher output throughput. Astra has the higher intelligence index. The Low and Medium labels are part of the conModel ComparisonClaude Haiku 5.5 (Xhigh) vs GPT-6 Astra (Medium)Haiku 5.5 Xhigh has lower listed prices than GPT-6 Astra Medium and a lower intelligence index. Haiku’s reported output throughput is higher, but Astra’s first-token delay is shorter. This pair illustrates why a reasoninModel ComparisonLing 3.1 Flash vs GPT-6 Astra (Medium)Ling 3.1 Flash has lower listed token prices, higher output throughput and a shorter first-token delay than GPT-6 Astra Medium in the recorded data. Astra has the higher intelligence index. The current source records con