heroku only shows up on python, for every model
Next.js
0of 29
heroku
- Claude0/18
- DeepSeek0/5
- GLM0/6
Python
14of 30
heroku
- Claude7/18
- DeepSeek3/6
- GLM4/6
Claude vs DeepSeek vs GLM: what coding agents pick
same prompts, claude code next to two free models in opencode. same top pick in 2 of 3 on python, in 0 of 3 on next.js.
ORM, deployment, email on next.js and fastapi. updated Oct 27, 2026
the two biggest splits
resend never on python
Next.js
16of 30
resend
- Claude11/18
- DeepSeek3/6
- GLM2/6
Python
0of 29
resend
- Claude0/18
- DeepSeek0/5
- GLM0/6
raw smtp 27/29
the three models
Claude Code
Opus 5.5, Sonnet 5.5, Haiku 5.5
- per cell
- 18
- scored
- 108/108
- cost
- $11.21
DeepSeek V4.1 Flash
via opencode
- per cell
- 6
- scored
- 34/36
- cost
- $0
GLM 5.3 Flash
via opencode
- per cell
- 6
- scored
- 35/36
- cost
- $0
top pick, by stack
splits highlighted
category, stackClaudeDeepSeekGLM
- ORMNext.jssplit12/18Drizzle ORM2/6tie Drizzle ORM / Prisma5/6Drizzle ORM
- ORMPythonsame18/18SQLAlchemy6/6SQLAlchemy4/5SQLAlchemy
- deploymentNext.jssplit14/18Plain Docker2/5tie Fly.io / Plain Docker3/6tie built its own / Plain Docker
- deploymentPythonsplit7/18tie Heroku / Plain Docker3/6Heroku4/6Heroku
- emailNext.jssplit11/18Resend3/6tie Raw SMTP / Resend4/6Raw SMTP
- emailPythonsame17/18Raw SMTP5/5Raw SMTP5/6Raw SMTP
every pick
one bar per model per stack. thin bars are single claude models.
ORM
- sqlalchemy
- drizzle orm
- hand-written sql
- kysely
- prisma
- built its own
- sqlmodel
Next.js + TypeScript
- Claudedrizzle orm 12/18
- opus 4/6
- sonnet 4/6
- haiku 4/6
- DeepSeekdrizzle orm / prisma 2/6
- GLMdrizzle orm 5/6
Python + FastAPI
- Claudesqlalchemy 18/18
- opus 6/6
- sonnet 6/6
- haiku 6/6
- DeepSeeksqlalchemy 6/6
- GLMsqlalchemy 4/5
deployment
- plain docker
- heroku
- fly.io
- built its own
Next.js + TypeScript
- Claudeplain docker 14/18
- opus 4/6
- sonnet 4/6
- haiku 6/6
- DeepSeekfly.io / plain docker 2/5
- GLMbuilt its own / plain docker 3/6
Python + FastAPI
- Claudeheroku / plain docker 7/18
- opus 4/6
- sonnet 3/6
- haiku 4/6
- DeepSeekheroku 3/6
- GLMheroku 4/6
- raw smtp
- resend
- postmark
- sendgrid
Next.js + TypeScript
- Clauderesend 11/18
- opus 4/6
- sonnet 4/6
- haiku 5/6
- DeepSeekraw smtp / resend 3/6
- GLMraw smtp 4/6
Python + FastAPI
- Clauderaw smtp 17/18
- opus 5/6
- sonnet 6/6
- haiku 6/6
- DeepSeekraw smtp 5/5
- GLMraw smtp 5/6
the fine print
- Claude per cell
- 18
- DeepSeek and GLM per cell
- 6
- step cap
- 12
- cost for these two
- $0
small free tier samples in opencode, read them as a lead. 3 of 72 free model runs made no attempt and are left out.
claude vs deepseek vs glm, questions
do claude, deepseek and glm pick the same tools?
same top pick in 2 of 6 category and stack pairs. 14 of 30 Python deploy runs picked heroku. Next.js 0 of 29. Next.js: resend 16 of 30. Python: resend 0, raw smtp 27 of 29.
which ORM do claude, deepseek and glm pick?
ORM on Next.js: Claude drizzle orm 12 of 18, DeepSeek drizzle orm / prisma 2 of 6, GLM drizzle orm 5 of 6. on Python: Claude sqlalchemy 18 of 18, DeepSeek sqlalchemy 6 of 6, GLM sqlalchemy 4 of 5.
which deployment do claude, deepseek and glm pick?
deployment on Next.js: Claude plain docker 14 of 18, DeepSeek fly.io / plain docker 2 of 5, GLM built its own / plain docker 3 of 6. on Python: Claude heroku / plain docker 7 of 18, DeepSeek heroku 3 of 6, GLM heroku 4 of 6.
which email do claude, deepseek and glm pick?
email on Next.js: Claude resend 11 of 18, DeepSeek raw smtp / resend 3 of 6, GLM raw smtp 4 of 6. on Python: Claude raw smtp 17 of 18, DeepSeek raw smtp 5 of 5, GLM raw smtp 5 of 6.
how many runs is this?
claude code: 18 runs per category per stack (opus 5.5, sonnet 5.5, haiku 5.5, 6 each). DeepSeek V4.1 Flash: 6, GLM 5.3 Flash: 6, run through OpenCode on the free tier. same prompts, same starters, same detector. small samples, read them as a lead.