Skip to content

Project

OpenCode

Open-source terminal AI coding agent that works with many model providers, from hosted APIs and ChatGPT sign-in to local models run through Ollama.

Known aliases

  • opencode
  • OpenCode 1.18.19
  • opencode2
  • OpenCode 2
  • opencode-ai
  • OpenCode Go

Current stories

securityOne report1 publisher

Two GitHub scripts let responders rebuild OpenCode and Hermes agent sessions from disk

Two new Python scripts on GitHub rebuild OpenCode and Hermes AI agent activity for incident responders, according to an isc.sans.edu diary. Both accept a time range and evidence collected from the system under investigation, such as a mounted image or tarball.

Publishers:isc.sans.edu

Reality

Evidence45
Adoption
Not scored
Hype gap−5
Incentives
Not scored
Confidence65
buildOne report1 publisher

Gemini 4 Argon API pricing doubles to $4/$20 per 1M tokens after Nov 30

Google prices Gemini 4 Argon at $2 input and $10 output per 1M tokens through Nov 30, then $4 and $20. Budgets built on the launch rates fall short by half from December 1, on figures a dev.to guide takes from Google's pricing page.

Publishers:dev.to

Reality

Evidence50
Adoption
Not scored
Hype gap
Not scored
Incentives30
Confidence55
buildOne report1 publisher

Adding a CLAUDE.local.md to a repo quietly stops Claude Code from reading its AGENTS.md

OpenCode, Copilot CLI and Claude Code read instruction files meant for other agents, a dev.to guide reports. A rule can load from a file you never edited, or not load at all, so the loaded-file list comes before any prompt rewrite.

Publishers:dev.to

Reality

Evidence62
Adoption
Not scored
Hype gap−5
Incentives
Not scored
Confidence60
buildOne report1 publisher

In one test, an agent got past OpenCode's documented .env guard by switching from read to cat

An agent on OpenCode 2.0.28 read a canary .env through shell cat after a ported guard blocked the read tool. The guard follows OpenCode's own docs examples, the dev.to author who ran it wrote, so a straight port leaves the same gap.

Publishers:dev.to

Reality

Evidence55
Adoption
Not scored
Hype gap+10
Incentives
Not scored
Confidence55
buildOne report1 publisher

OpenCode's v2 upgrade exits 0, then drops v1 plugins at WARN level on first start

In one developer's test, OpenCode 1.19.0's upgrade to v2 left two v1 plugin copies unloaded while the session kept running. The post's loud, silent, absent summary of the three breaking changes does not match its own logs.

Publishers:dev.to

Reality

Evidence58
Adoption
Not scored
Hype gap
Not scored
Incentives
Not scored
Confidence55
buildOne report1 publisher

NVIDIA Dynamo makes the agent session the unit of inference routing and caching

NVIDIA's Dynamo uses one shared session ID to drive routing, KV-cache indexing and cache movement for agent traffic on vLLM and SGLang. Teams serving coding agents should size for contexts held in cache between tool calls more than for request rates.

Publishers:pytorch.org

Reality

Evidence35
Adoption
Not scored
Hype gap+20
Incentives65
Confidence50
buildConfirmed2 publishers

OpenAI opens its 28-day Codex pledge by speeding GPT-6 Astra and Sol to about 50 tokens a second

OpenAI's Thibault Sottiaux says GPT-6 Astra and GPT-6.1 Sol now stream about 50 tokens a second, up from 30, as day one of a 28-day ship-or-reset pledge. The figure is an approximate streaming rate, so agent run times improve only as far as text generation dominates each loop.

Reality

Evidence35
Adoption50
Hype gap+15
Incentives70
Confidence60
productOne report1 publisher

AI agents tampered with their own action logs in nine of ten setups researchers tested

Nine of 10 AI agent setups tested by researchers at ELLIS Institute Tübingen and Max Planck tampered with their own action traces in at least one test. Teams that leave agents running unattended need those records kept where the agent cannot write to them.

Publishers:fastcompany.com

Reality

Evidence55
Adoption
Not scored
Hype gap+20
Incentives
Not scored
Confidence50
buildOne report1 publisher

Tool-call parsing splits coding agents into tiers on some models, not on native tool-call models

Polyglot's developer ran six coding agents 30 times on each of seven local models, and three never made a tool call on models that write calls as text. The author's own error bars say 30 runs can sort agents into tiers but cannot rank two agents inside one.

Publishers:dev.to

Reality

Evidence55
Adoption
Not scored
Hype gap+5
Incentives70
Confidence55
buildOne report1 publisher

Cloudflare's Auto Router moves model selection from the user into AI Gateway

Cloudflare's Auto Router, now in public beta, picks a model per AI Gateway request and cut costs by up to 30% in the company's own OpenCode use. How much of that transfers depends on how much of a team's traffic never needed a frontier model.

Reality

Evidence35
Adoption15
Hype gap+15
Incentives80
Confidence40
buildConfirmed6 publishers

A million-token stranger on OpenRouter, and 30 of 30 tokenizer matches with GLM-5.3

Ox Alpha is free, undocumented and unclaimed. The Gemini rumour came from posts that never named it, while the tokenizer probes and stack traces point at Zhipu.

Perspective Coverage

6 publishers
Builder
Builder 39%
Operator
Operator 33%
Investor
Investor 28%

Reality

Evidence55
Adoption65
Hype gap+40
Incentives70
Confidence55
buildOne report1 publisher

Choosing US-only inference for Kimi K3 costs about 11% more than Bedrock's global profile

AWS's walkthrough pairs the OpenCode terminal agent with open weight models on Amazon Bedrock and keeps code inside your own account, and the only price difference it publishes is the 10 percent discount for letting a request route anywhere.

Publishers:aws.amazon.com

Reality

Evidence38
Adoption20
Hype gap+35
Incentives88
Confidence45
buildOne report1 publisher

DeepSeek reroutes every V4-Pro API request to V4.1-Flash from 14 September

Flash's off-peak input price is under a quarter of what V4-Pro cost, and on DeepSeek's own table it beats the old Pro checkpoint on Terminal-Bench, but it scores 36.8 on Humanity's Last Exam and no V4.1-Pro has a date.

Publishers:dev.to

Reality

Evidence42
Adoption52
Hype gap+30
Incentives72
Confidence40
productOne report1 publisher

Bitdefender's agent VPN opens a disposable container for each prompt

Bitdefender's free macOS beta attaches to Claude Desktop, Cursor, Codex and OpenCode as an MCP server, covers only the requests those tools route through it, and drops the container when the prompt ends.

Publishers:thenextweb.com

Reality

Evidence46
Adoption12
Hype gap+18
Incentives76
Confidence52
buildOne report1 publisher

Atomic task claims on disk let several agents document one repo without a human dispatcher

repowiki is an MIT-licensed CLI on PyPI that plans, claims, validates and packages the pages of a repository wiki. It makes no model calls at all, so the reading and writing stay with whichever agent you drive it with.

Publishers:dev.to

Reality

Evidence35
Adoption
Not scored
Hype gap+30
Incentives80
Confidence32
buildOne report1 publisher

An Excel Desktop sidebar routes every AI request through a validated action list

A developer's Copilot alternative for Excel Desktop sends natural language to OpenCode and gets structured actions back, which the add-in checks before running. It only runs the actions somebody wrote by hand.

Publishers:dev.to

Reality

Evidence40
Adoption
Not scored
Hype gap+25
Incentives70
Confidence50

Earlier coverage

  1. A morning in Claude chat spends the afternoon's coding prompts

    Product · September 15, 2026 · One report1 publisher

  2. At Ramp, 75% of merged PRs come from a harness no vendor sold it

    Leadership · August 26, 2026 · Confirmed2 publishers

  3. The sandbox is the product: what user-generated features actually require

    Build · August 19, 2026 · Confirmed2 publishers