Engineering Blog

Notes from the build

Engineering insights, architecture deep-dives, and product notes from the Vesprr Software team.

#AI models

Uzair bin HaroonNov 11, 2026

What is Claude Fable 5.2? Anthropic's model for the hardest work

Claude Fable 5.2 is Anthropic's top model for long, hard technical work, out 9 Nov 2026. It costs $8/$40 per million tokens, twice Opus 5.5.

AI modelsClaude Fable 5.2claude-fable-5-2

Sadaf MasoodNov 4, 2026

Kimi K3.1 Pro pricing: $2 in, $10 out, and what a job costs

Kimi K3.1 Pro costs $2 per million input tokens, $10 per million output and $0.20 cached. A cached 1,000-question day costs about $31.

AI modelsKimi K3.1 ProKimi K3.1 Pro pricing

Zohaib MasoodNov 1, 2026

Why you're using Jev wrong: four fixes from TypeSafe's docs

Most bad Jev results come from an unpinned alias, maths left to the model, vague questions and one threshold for every action. TypeSafe's docs give the fixes.

AI modelsJevJev limitations

Zohaib MasoodOct 28, 2026

What is GPT-6.1 Terra? OpenAI's $1 model for images and screens

GPT-6.1 Terra is OpenAI's mid-priced model for images, screens and writing, out 26 Oct 2026: $1 in, $5 out per 1M tokens, in the API and ChatGPT Work.

AI modelsGPT-6.1 Terragpt-6.1-terra

Amir KhanOct 25, 2026

What is Kimi K3.1 Pro? Moonshot AI's new flagship model

Moonshot AI's flagship model, live on the Kimi API since 23 Oct 2026: 1M-token context, $2/$10 per 1M tokens, open weights on Hugging Face.

AI modelsKimi K3.1 Prokimi-k3.1-pro

Sadaf MasoodOct 19, 2026

What is Gemini 4 Argon? Google's model for long, hard tasks

Google's flagship reasoning model, GA 13 Oct 2026, for long coding and agent work. $2/$10 per 1M tokens to 30 Nov, then $4/$20.

AI modelsGemini 4 Argongemini-4-argon

Uzair bin HaroonOct 18, 2026

What is Muse Wonder? Meta's frontier model for agents

Muse Wonder is Meta's top model for coding and long agent tasks, released 12 Oct 2026 at $4/$20 per 1M tokens, the same as Claude Opus 5.5. It is in preview.

AI modelsMuse WonderMeta Muse Wonder

Zohaib MasoodOct 10, 2026

What is Claude Haiku 5.5? Anthropic's $0.10 small model

Anthropic's small model for high-volume work: classification, extraction, support, subagents. $0.10/$0.50 per 1M tokens, 90% below Haiku 4.5.

AI modelsClaude Haiku 5.5Claude Haiku 5.5 pricing

Uzair bin HaroonOct 2, 2026

What is Cloudflare Clef? A decision model for your workflows

Clef returns typed decisions from text or images. Try it for routing and classification through Workers AI, or run the open weights yourself.

AI modelsClefCloudflare Clef

Sadaf MasoodSep 25, 2026

Claude Opus 5.5 pricing: $4 in, $20 out, and what a job costs

Opus 5.5 costs $4 per million input tokens and $20 per million output, with thinking billed as output. A 10,000-document job runs about $390 a month.

AI modelsClaude Opus 5.5Claude Opus 5.5 pricing

Uzair bin HaroonSep 24, 2026

What is Jev? TypeSafe AI's model that returns types, not text

Jev returns a typed value and its probabilities instead of a sentence, at $0.042 per million input tokens. What it does, what it gets wrong, who it suits.

AI modelsJevTypeSafe AI

Sadaf MasoodSep 16, 2026

Jev pricing: $0.042 per million input tokens, output free

TypeSafe AI's Jev charges $0.042 per million input tokens and nothing for output. A hundred thousand documents classified for $8.40, and what is unproven.

AI modelsJevTypeSafe AI