τTaosisTaoScale

Bittensor subnet news

Subnet teams' own channels, X posts, press, newsletters and events read from the chain — GitHub lives in its own tab

Lower prices on Terza Inference 🔻 New Standard-tier capacity for GPT-6 models, ~33% cheaper than before: • gpt-6.1-sol / gpt-6-sol: $0.71 in / $3.53 out per 1M • gpt-6-astra: $3.53 / $17.65 • gpt-6.1-terra: $0.35 / $1.76 Image input supported. Economy tier still from $0.09/1M. Served by miners on the Terza subnet (SN103) ⛏️

Full post on TaosisOpen on X ↗

GPT Image 2.6 Flare and Sunburst are live on Terza Inference, 29% below OpenAI's list price. A 1024×1024 image costs $0.018 at low, $0.031 at medium and $0.038 at high. Flare is built for fast drafts, Sunburst for detailed final images. Same OpenAI-compatible API, served by miners on Bittensor SN103

Full post on TaosisOpen on X ↗

Taosis newsletters

Taosis News: the week’s articles from the subnet teams, the press and community newsletters, every Friday. Taosis Weekly: the week’s movers, best yields and chain events, with dollar values, every Monday. Past issues

Newsletters

Your address is used only to send the newsletters you tick, with your consent (double opt-in). Unsubscribe anytime · Privacy

On Taosis

Swap TAO ⇄ USDC

USDC on Ethereum to TAO, or back. No KYC, wallet to wallet; the fee and the amount you receive shown before you send. Not instant: run by Taosis, paid by hand within hours.

Open →
On-chainChild keys103SN1036d ago

Validator FirstTensor.com (5DvT…N2ed) scheduled a child-key change on 42 subnets

Validator hotkey 5DvTpiniW9s3APmHRYn8FroUWyfnLtrsid5Mtn5EwMXHN2ed announced new child keys on SN6, SN8, SN9, SN14, SN20, SN29, SN30, SN35, SN37, SN38, SN41, SN44 and 30 more, applying at block 9,431,952 (~24h). Stake delegated to it there routes through its children after that.

ValidatorFirstTensor.comSubnetsSN6, SN8, SN9, SN14, SN20, SN29, SN30, SN35, SN37, SN38, SN41, SN44, … (42 subnets)

Between blocks 9,424,713 and 9,424,772

Terza Inference now accepts images on text models. Send screenshots and photos to Claude, GPT, Gemini, DeepSeek, GLM, Kimi and Qwen through the same API, in both OpenAI and Anthropic formats. Works in Claude Code too. Models that can't read images now return a clear error. No charge for rejected requests.

Full post on TaosisOpen on X ↗

Context limits lifted on Terza Inference. Claude and GPT now take up to 1M input tokens and return up to 128K output tokens, with long responses streaming until they’re done. DeepSeek, Kimi, GLM, Qwen and Gemini follow over the next few hours. If you hit max_tokens or input too long errors last week, that's fixed. Thanks for the reports. 🙏

Full post on TaosisOpen on X ↗

Long outputs are live on Terza Inference. max_tokens now goes up to 128K (131,072) on Claude and GPT models, and long responses stream for as long as they need. No more cut-offs mid-answer. Built for coding agents, long docs and big refactors.

Full post on TaosisOpen on X ↗