Compare models

Put up to 4 models beside each other — token prices, context windows, capabilities and provider, from the same catalogue the model pages read.

GPT-6 Sol ProOpenAI Remove
Gemini 3.5 TranscribeGoogle Remove
Grok 4.7SpaceXAI Remove
Aura-2Deepgram Remove

4 is the maximum. Remove one to add another.

gpt-6-sol-pro vs gemini-3.5-transcribe vs grok-4.7 vs aura-2
Attribute	GPT-6 Sol Progpt-6-sol-pro	Gemini 3.5 Transcribegemini-3.5-transcribe	Grok 4.7grok-4.7	Aura-2aura-2
Pricing
Input	$2.00 / 1M	$0 / 1M	$1.60 / 1M	$0 / 1M
Output	$10.00 / 1M	$0 / 1M	$4.80 / 1M	$0 / 1M
Cache Write (5m)	$2.00 / 1M	Not applicable	$1.60 / 1M	Not applicable
Cache Write (1h)	$2.00 / 1M	Not applicable	$1.60 / 1M	Not applicable
Cache Read	$2.00 / 1M	Not applicable	$1.60 / 1M	Not applicable
Web Search	$0 / 1M	$0 / 1M	$0 / 1M	$0 / 1M
Context
Max context	1.1M	98.3K	500K	N/A
Max output	N/A	N/A	N/A	N/A
Capabilities
Vision	Yes	Yes	Yes	No
Function Calling	Yes	Yes	Yes	No
JSON Mode	Yes	No	Yes	No
Streaming	Yes	No	Yes	No
Catalogue
Provider	OpenAI	Google	SpaceXAI	Deepgram
Category	chat	voice	chat	voice
Charge type	Pay As You Go	Pay As You Go	Pay As You Go	Pay As You Go
Released	—	2026-09-25	—	2026-07-16
Description
Summary	GPT-6 Sol Pro uses the same underlying model as GPT-6 Sol, but runs with reasoning.mode set to pro for higher-quality responses on complex and demanding tasks. Optimized for deeper reasoning, greater accuracy, and more reliable multi-step execution, it is particularly well suited for agentic coding, long-horizon software engineering, professional analysis, and complex automated workflows where solution quality takes priority over latency and cost.	Google Gemini 3.5 Transcribe speech-to-text. Billed per input and output token.	Grok 4.7 is SpaceXAI's flagship model for coding, agentic workflows, and professional knowledge work, succeeding Grok 4.6. It is particularly strong at long-running software engineering, self-verification, and long-context execution, while improving capabilities in document drafting, presentations, and other professional tasks. Trained with extended reinforcement learning focused on multi-hour problems, Grok 4.7 is optimized for sustained, complex task execution and natively supports the Grok Bot harness for conversational workflows. It also introduces an enhanced safeguard stack designed to combine strong jailbreak resistance with low refusal rates for legitimate technical work. Reported benchmark results use xhigh reasoning effort.	Deepgram Aura-2 text-to-speech for real-time voice applications. Billed per input character.