MindsDB

New model GPT-6.1 Terra (OpenAI) added to the model price table at $1.00 input, $5.00 output, $0.05 cached input, $1.25 cache write per 1M tokens.

Two new models added to the pricing table: Kimi K3.1 Pro (Moonshot AI, $2.00 in/$10.00 out per 1M tokens) and DeepSeek V4.1 Flash 1017 (Fireworks AI, $0.30/$1.20).

Two models added to the token price table: Gemini 4 Argon (Google, $2.00 in/$10.00 out per 1M) and Muse Wonder 1.0 (Meta, $4.00 in/$20.00 out per 1M).

Weekly roll-up · includes Oct 16 daily, Oct 15 daily

Added Gemini 4 Argon (Google) to the model rate card: $2.00 input, $10.00 output, $0.10 cached input, no cache-write charge.

New model added to rate card: Muse Wonder 1.0 (Meta) at $4.00 input, $20.00 output, $0.20 cache read, $4.00 cache write per 1M tokens

Pricing page moved from a plan-less rate card ('no subscription or plan') to explicit Free ($0/month) and Pay as you go (prepaid balance, provider rates) plan cards; model rates unchanged.

DeepSeek V4.1 Flash inference rates raised ($0.22/$0.66 to $0.30/$1.20 per 1M in/out); MindsHub Foundry 100% credit match promo (up to $5,000) removed

Weekly roll-up · includes Oct 3 daily, Oct 2 daily

MindsHub Foundry promo removed: 100% match on inference-credit purchases for first three months, up to $5,000, for selected teams.

Claude Sonnet 5.5 (Anthropic) added to the rate card: input $2.00, output $10.00, cached input $0.20, cache write $2.50 per 1M tokens. DeepSeek V4.1 Flash (Fireworks AI) token rates raised: input $0.22 to $0.30, output $0.66 to $1.20, cache write $0.22 to $0.30; cached input cut $0.01 to $0.006.

Added free model Jev (classification/routing) and paid models Claude Opus 5.5, GPT-6 Sol, GPT-6 Luna, Grok 4.7 to the rate card; removed DeepSeek V4 Pro and DeepSeek V4 Flash.

Weekly roll-up · includes Sep 24 daily

Free tier expanded to include Jev, TypeSafe's decision model for classification, routing and scoring, alongside the existing MindsHub Air model, both usable at no charge under fair use

MindsHub Blaze (Cerebras) rates increased: input $0.35 to $0.99, output $0.75 to $1.49, cached input $0.35 to $0.99 per 1M tokens Free tier restructured from a metered 5,000,000 token/month allowance on MindsHub Air to an unmetered 'free within fair use' tier with daily rate limits and a fair-use policy

Weekly roll-up · includes Sep 15 daily

MindsHub Blaze (Cerebras) rates rose (input $0.35→$0.99, output $0.75→$1.49, cached $0.35→$0.99/1M tokens); Air's free tier switched from a fixed 5M tokens/month cap to open-ended fair-use limits.

Two new models added to the model rate card: MindsHub Blaze (Cerebras, $0.35 input/$0.75 output/$0.35 cached per 1M tokens, free cache writes) and GPT-OSS 120B (Fireworks AI, $0.15 input/$0.60 output/$0.02 cached/$0.15 cache-write per 1M tokens).

MindsHub Blaze (Cerebras), Claude Fable 5.1, GPT-6 Astra, and Gemini 3.8 Flash each added as new models on the token-based rate card, each with its own per-model price

Weekly roll-up · includes Sep 14 daily, Sep 8 daily

Rate card added 3 models: DeepSeek V4 Flash ($0.22/$0.66/$0.01/$0.22), GLM 5.3 ($1.40/$4.40/$0.26/$1.40), GLM 5.3 Flash ($0.15/$0.50/$0.03/$0.15). Separately, two identically-named DeepSeek V4-Pro-0813 rows were renamed to DeepSeek V4 Pro and DeepSeek V4 Pro (0813 pin); the pin variant repriced from $1.74/$3.48/$0.15/$1.74 to $1.32/$3.96/$0.05/$1.32.

MindsHub scrubbed every mention of the 5% platform fee: hero bullet, the 'Platform fee' info box, the volume-pricing note, and the FAQ answer all dropped the fee language. Rate table is unchanged.

MindsHub split Free/Pro into Cowork plus a new Unified Inference API product; both add a 5% platform fee on rate-card prices. Gemini 3.6 Flash rates were halved; Kimi K3 gained a cache-write fee.

Pro plan billing mechanism changed from postpaid spend controls (choose top-up cadence, set budget limits, cap spend per person/team) to a prepaid balance model (add credit when you want, or auto-recharge with a monthly charge cap)

Community support (Discord) added to Free plan; Support tickets (open a ticket, tracked reply) added to Pro plan. Free plan dropped web-app support — Cowork is now macOS and Windows only (previously macOS, Windows and the web).