HASTATAgent stack guide

Agent stack recommender

Choose the stack, not just the model.

Match your harness, workload, billing constraints, context, and budget to a practical setup you can run now.

Recommendation beta

Live provider offers joined to dated editorial evidence; never a universal best-model score.

Read the methodology →

Your constraints

API token amounts are uncached planning scenarios, not spend predictions or subscription quotas. Subscription activity profiles remain separate; cache discounts are never applied.

Recommendation results

DeepSeek V4 Pro

Fresh · fetched 2026-08-17T02:11:57.739Z · Models.dev

Best API stack

Editorial evidence rank 2 · independent

A reviewed second-tier everyday candidate where the direct offer can be cost-effective.

Published ranking rule: Balanced: ranks through bestRank + 1; then lower uncached estimate, evidence class (independent, mixed, official-only), editorial rank, direct provider, and stable provider/model/evidence IDs.

Primary model
DeepSeek V4 ProDeepSeek (deepseek) · 1M context$0.43 input$0.87 output/ 1M$0.004 cache read / 1M
Planning estimate
$5.22 / month10M input + 1M output / month · uncached input + output; not a spend prediction
Budget fit
Within monthly budget $100
Background role
DeepSeek V4 FlashDeepSeek (deepseek) · $0.14 input · $0.28 output / 1M · $1.68 full-volume estimate
Connection facts
DeepSeek · deepseek-v4-proProvider documentation · Direct API route; dsh is a developer preview and is limited to this route.

Background uses DeepSeek V4 Flash via DeepSeek; its full-volume estimate is cheaper than the primary. No image-capable evidence-matched vision offer passed the same-provider gates.

Evidence sources (2)

Caveats

  • Model evaluations do not prove performance in the selected harness.
  • Only exact current and 0813 aliases match.

Eligible alternatives

Qwen3.8 Max (hosted)Alibaba (alibaba) · 1M context$2.00 input · $6.00 output / 1M$26 · Editorial evidence rank 2Higher uncached estimate within the allowed editorial rank window.
Kimi K3OpenRouter (openrouter) · 1.05M context$3.00 input · $15.00 output / 1M$45 · Editorial evidence rank 2Higher uncached estimate within the allowed editorial rank window.
Claude Opus 4.7OpenRouter (openrouter) · 1M context$5.00 input · $25.00 output / 1M$75 · Editorial evidence rank 2Higher uncached estimate within the allowed editorial rank window.

Curated subscription option

verified

QwenCloud Token Plan Individual Lite

QwenCloud Token Plan Individual Lite is a personal interactive plan, not a general API allowance.

QwenCloud Token Plan Individual Lite is a curated route for everyday interactive coding.

Actual monthly price
$6/month$6/month at the reviewed date; 2,500 credits per seven-day window. Prices and quota windows can change.
Plan type
Quota / window
Model scope
Hosted qwen3.8-max plus the current models listed by the individual plan.1M reviewed recommendation context
Key limits
2,500 model credits / week

Caveats

  • Individual plans are interactive-only, single-device personal use; backend and batch automation are prohibited.
  • Use can involve Singapore/global data transfer.
  • Seven-day quota windows and current prices can change; credits are not API tokens.
  • The reviewed primary model is qwen3.8-max; the official table also lists current DeepSeek and GLM routes.

Official subscription source·Official subscription source·Official subscription source

Subscription credits, quotas, windows, and API token scenarios are separate units. Nominal vendor credits are never compared across providers.

Live decision data

Inspect the model and provider catalogue.

Compare distinct models without losing provider-specific offers, prices, and caveats.

ModelsProviders