Inkling 2 Small vs Mistral Large 4 High compared

This comparison puts Thinking Machines's Inkling 2 Small next to Mistral's Mistral Large 4 (High). Cost is where they split the most, with Inkling 2 Small working out at about 45% of Mistral Large 4 High's blended rate. What follows is built entirely from tracked data, so where the catalogue is silent, this page says so instead of guessing.

Price per token

Inkling 2 Small lists $0.30 per million input tokens and $1.00 per million output tokens; Mistral Large 4 High lists $0.70 and $2.10. Mistral Large 4 High costs 2.1 times as much per output token.

A reference job of 1 million input tokens and 200,000 output tokens costs $0.50 with Inkling 2 Small and $1.12 with Mistral Large 4 High, so Inkling 2 Small saves $0.62. A bigger example, an agent that reads 50 million tokens and writes 5 million in a month, costs $20.00 on Inkling 2 Small and $45.50 on Mistral Large 4 High: a gap of $25.50. Blended at three input tokens per output token, the effective rates are $0.475 for Inkling 2 Small and $1.05 for Mistral Large 4 High.

Release timing

Inkling 2 Small arrived on November 5, 2026, 7 days after Mistral Large 4 High (October 29, 2026). That is close enough to count them as the same generation.

The headline benchmark

Inkling 2 Small records 30.6 on the intelligence index and Mistral Large 4 High records 26.1. It is a meaningful gap rather than a rounding difference. Put another way, Mistral Large 4 High would need 17% more on this index to draw level.

Across the 109 active models with an intelligence score in the catalogue, Inkling 2 Small ranks 52nd and Mistral Large 4 High ranks 65th.

For coding or maths work specifically, the catalogue has no dedicated index on either side yet.

Value for money

A blunt value measure, blended price over intelligence score, puts Inkling 2 Small at $0.016 and Mistral Large 4 High at $0.040 per point. Inkling 2 Small scores higher and costs less per point, so on price per point there is nothing to trade off.

Throughput and time to first token

Measured output speed is 139 tokens per second for Inkling 2 Small and 139 tokens per second for Mistral Large 4 High. There is little to choose between them on raw output speed. Before any text appears, Inkling 2 Small takes 2.2 seconds and Mistral Large 4 High 1.7 seconds. In an interactive app, Mistral Large 4 High will feel noticeably more responsive.

Put together, a 600-token answer would take about 6.6 seconds from Inkling 2 Small and 6.0 seconds from Mistral Large 4 High, counting the first-token wait.