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.

Voxtral Mini 3B 2507Mistral AI Remove
Grok 4.7SpaceXAI Remove
Perceptron Mk1.5Perceptron Remove
Qwen3 ASR 0.6BQwen Remove

4 is the maximum. Remove one to add another.

voxtral-mini-3b-2507 vs grok-4.7 vs perceptron-mk1.5 vs qwen3-asr-0.6b
Attribute	Voxtral Mini 3B 2507voxtral-mini-3b-2507	Grok 4.7grok-4.7	Perceptron Mk1.5perceptron-mk1.5	Qwen3 ASR 0.6Bqwen3-asr-0.6b
Pricing
Input	$0 / 1M	$1.60 / 1M	$0.15 / 1M	$0 / 1M
Output	$0 / 1M	$4.80 / 1M	$1.50 / 1M	$0 / 1M
Cache Write (5m)	Not applicable	$1.60 / 1M	$0.15 / 1M	Not applicable
Cache Write (1h)	Not applicable	$1.60 / 1M	$0.15 / 1M	Not applicable
Cache Read	Not applicable	$1.60 / 1M	$0.15 / 1M	Not applicable
Web Search	$0 / 1M	$0 / 1M	$0 / 1M	$0 / 1M
Context
Max context	N/A	500K	36.9K	N/A
Max output	N/A	N/A	N/A	N/A
Capabilities
Vision	No	Yes	No	No
Function Calling	No	Yes	No	No
JSON Mode	No	Yes	No	No
Streaming	No	Yes	Yes	No
Catalogue
Provider	Mistral AI	SpaceXAI	Perceptron	Qwen
Category	voice	chat	chat	voice
Charge type	Pay As You Go	Pay As You Go	Pay As You Go	Pay As You Go
Released	2026-08-13	—	2026-09-25	2026-08-13
Description
Summary	Mistral AI Voxtral Mini (3B) speech-to-text. Billed per second of audio.	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.	Perceptron Mk1.5 chat model that accepts audio input. Audio and text input are billed per token.	Qwen3-ASR 0.6B speech-to-text. Billed per second of audio.