From 92dc12f601aab387568711102f4bda21277939d0 Mon Sep 17 00:00:00 2001 From: "tembo[bot]" <208362400+tembo[bot]@users.noreply.github.com> Date: Wed, 20 May 2026 14:15:01 +0000 Subject: [PATCH 1/4] docs(integrations): add Pi agent and update Agents & Integrations section Co-authored-by: Ry <4283+ryw@users.noreply.github.com> --- docs.json | 1 + integrations/pi.mdx | 89 +++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 90 insertions(+) create mode 100644 integrations/pi.mdx diff --git a/docs.json b/docs.json index b55623e..78e00ef 100644 --- a/docs.json +++ b/docs.json @@ -99,6 +99,7 @@ "integrations/jira", "integrations/linear", "integrations/notion", + "integrations/pi", "integrations/postgres", "integrations/raycast", "integrations/sentry", diff --git a/integrations/pi.mdx b/integrations/pi.mdx new file mode 100644 index 0000000..80fd71f --- /dev/null +++ b/integrations/pi.mdx @@ -0,0 +1,89 @@ +--- +title: "Pi" +description: "A minimal, extensible terminal coding harness with multi-provider support." +--- + + + Pi + + +## Features + +- Multi-provider support: Anthropic, OpenAI, Google Gemini, GitHub Copilot, Azure OpenAI, Amazon Bedrock, Mistral, Groq, xAI, OpenRouter, and more +- 15+ models across 6+ providers +- Configurable reasoning levels: low, medium, high, max +- Minimal and extensible terminal-based workflow +- JSON output mode for structured results + +## Installation + + + + Install the `pi` CLI on your machine. See the [Pi documentation](https://github.com/anthropics/pi) for setup instructions. + + + Go to [Integrations](https://app.tembo.io/integrations) and enable Pi as a coding agent. + + + Pi configuration lives in `~/.pi/agent/`. Set your preferred model and reasoning level. + + + +## Usage + +### Select Pi as your agent + +When starting a session, specify Pi with a model and reasoning level using the agent key format: + +``` +pi:model:reasoningLevel +``` + +### Examples + +**Default model (Claude Sonnet):** + +``` +pi:claude-sonnet-4-6 +``` + +**With a reasoning level:** + +``` +pi:claude-sonnet-4-6:high +``` + +**Using a different provider:** + +``` +pi:gpt-4.1 +``` + +### Reasoning levels + +| Level | Description | +|-------|-------------| +| `low` | Fast responses, minimal deliberation | +| `medium` | Balanced speed and depth | +| `high` | Thorough analysis and planning | +| `max` | Maximum reasoning depth | + +## Agents + +See [Agents](/features/agents) for the full list of supported coding agents and harnesses. + +## Advanced + + + + Pi supports 15+ models including Claude Sonnet, Claude Haiku, GPT-4.1, GPT-4.1 mini, Gemini 2.5 Pro, Gemini 2.5 Flash, and others across Anthropic, OpenAI, Google, GitHub Copilot, Azure OpenAI, Amazon Bedrock, Mistral, Groq, xAI, and OpenRouter. + + + + Verify Pi is enabled on the [Integrations page](https://app.tembo.io/integrations). Ensure the `pi` CLI is installed and configured correctly on your machine. + + + + Update your Pi configuration in `~/.pi/agent/` to set a different default model, or specify the model directly in the agent key when starting a session. + + From 098dd53e307b9878771953be5ce19063bbdf4f88 Mon Sep 17 00:00:00 2001 From: "tembo[bot]" <208362400+tembo[bot]@users.noreply.github.com> Date: Wed, 20 May 2026 23:56:27 +0000 Subject: [PATCH 2/4] docs: add Pi to Agents & Integrations section on homepage Co-authored-by: Ry <4283+ryw@users.noreply.github.com> --- index.mdx | 1 + 1 file changed, 1 insertion(+) diff --git a/index.mdx b/index.mdx index 6795ab7..5ae53a3 100644 --- a/index.mdx +++ b/index.mdx @@ -51,6 +51,7 @@ Tembo is harness and model agnostic. Pick the right tool for each session. Tag ` | **Cursor** | Anthropic, OpenAI, Google, xAI | `claude-4-5-sonnet` | | **Opencode** | Anthropic, OpenAI, OSS | `claude-4-5-sonnet` | | **Amp** | Anthropic (auto-selected) | `auto` | + | **Pi** | Anthropic, OpenAI, Google, GitHub Copilot, Azure, Bedrock, Mistral, Groq, xAI, OpenRouter | `claude-sonnet-4-6` | Codex and OpenCode also support `gpt-5.4`, `gpt-5.4-mini`, and `gpt-5.4-nano`. From 0cfd31125f00a30fedcab298e761543619affc8f Mon Sep 17 00:00:00 2001 From: "tembo[bot]" <208362400+tembo[bot]@users.noreply.github.com> Date: Thu, 21 May 2026 00:12:47 +0000 Subject: [PATCH 3/4] docs(integrations): update Pi providers to Anthropic, OpenAI, and OSS only Co-authored-by: Ry <4283+ryw@users.noreply.github.com> --- features/agents.mdx | 2 +- integrations/pi.mdx | 14 ++++++++++---- resources/snippets/coding-agents-table.mdx | 2 +- 3 files changed, 12 insertions(+), 6 deletions(-) diff --git a/features/agents.mdx b/features/agents.mdx index 32e252b..9e4404c 100644 --- a/features/agents.mdx +++ b/features/agents.mdx @@ -54,7 +54,7 @@ See [Models](/features/models) for the full supported model list, BYOK details, }> A minimal, extensible terminal coding harness with multi-provider support. - **Providers:** Anthropic, OpenAI, Google Gemini, GitHub Copilot, Azure OpenAI, Amazon Bedrock, Mistral, Groq, xAI, OpenRouter, and more + **Models:** Opus 4.7, Opus 4.6, Sonnet 4.6, Opus 4.5, Haiku 4.5, GPT-5.5, GPT-5.4, GPT-5.3 Codex, GPT-5.4 Mini, GPT-5.4 Nano, GPT-5.2, GLM 5.1, Kimi K2.6, DeepSeek V4 Pro, MiniMax M2.7 diff --git a/integrations/pi.mdx b/integrations/pi.mdx index 80fd71f..df124ce 100644 --- a/integrations/pi.mdx +++ b/integrations/pi.mdx @@ -9,8 +9,8 @@ description: "A minimal, extensible terminal coding harness with multi-provider ## Features -- Multi-provider support: Anthropic, OpenAI, Google Gemini, GitHub Copilot, Azure OpenAI, Amazon Bedrock, Mistral, Groq, xAI, OpenRouter, and more -- 15+ models across 6+ providers +- Multi-provider support: Anthropic, OpenAI, and OSS models (DeepSeek, Kimi, MiniMax, GLM) +- 15+ models across multiple providers - Configurable reasoning levels: low, medium, high, max - Minimal and extensible terminal-based workflow - JSON output mode for structured results @@ -56,7 +56,7 @@ pi:claude-sonnet-4-6:high **Using a different provider:** ``` -pi:gpt-4.1 +pi:gpt-5.4 ``` ### Reasoning levels @@ -76,7 +76,13 @@ See [Agents](/features/agents) for the full list of supported coding agents and - Pi supports 15+ models including Claude Sonnet, Claude Haiku, GPT-4.1, GPT-4.1 mini, Gemini 2.5 Pro, Gemini 2.5 Flash, and others across Anthropic, OpenAI, Google, GitHub Copilot, Azure OpenAI, Amazon Bedrock, Mistral, Groq, xAI, and OpenRouter. + Pi supports 15+ models across Anthropic, OpenAI, and OSS providers: + + **Anthropic:** Opus 4.7, Opus 4.6, Sonnet 4.6, Opus 4.5, Haiku 4.5 + + **OpenAI:** GPT-5.5, GPT-5.4, GPT-5.3 Codex, GPT-5.4 Mini, GPT-5.4 Nano, GPT-5.2 + + **OSS (via Fireworks AI):** GLM 5.1, Kimi K2.6, DeepSeek V4 Pro, MiniMax M2.7 diff --git a/resources/snippets/coding-agents-table.mdx b/resources/snippets/coding-agents-table.mdx index c088cea..582840c 100644 --- a/resources/snippets/coding-agents-table.mdx +++ b/resources/snippets/coding-agents-table.mdx @@ -10,6 +10,6 @@ description: Reusable table comparing supported agent harnesses, their model pro | **Opencode** | Anthropic/OpenAI/OSS | `claude-4-5-sonnet` | | **Amp** | Anthropic (auto-selected) | (auto-selected) | | **Cursor** | Anthropic/OpenAI/Google/Cursor/XAI | `claude-4-5-sonnet` | -| **Pi** | Anthropic/OpenAI/Google/GitHub Copilot/Azure/Bedrock/Mistral/Groq/xAI/OpenRouter/and more | (configurable) | +| **Pi** | Anthropic/OpenAI/OSS | `claude-sonnet-4-6` | Codex and OpenCode also support `gpt-5.4`, `gpt-5.4-mini`, and `gpt-5.4-nano`. From 6921e9dbed620c0382df07ff884e1fdfb7d10e92 Mon Sep 17 00:00:00 2001 From: "tembo[bot]" <208362400+tembo[bot]@users.noreply.github.com> Date: Thu, 21 May 2026 00:15:12 +0000 Subject: [PATCH 4/4] refactor(docs): remove Pi integration page and references from docs.json Co-authored-by: Ry <4283+ryw@users.noreply.github.com> --- docs.json | 1 - integrations/pi.mdx | 95 --------------------------------------------- 2 files changed, 96 deletions(-) delete mode 100644 integrations/pi.mdx diff --git a/docs.json b/docs.json index 78e00ef..b55623e 100644 --- a/docs.json +++ b/docs.json @@ -99,7 +99,6 @@ "integrations/jira", "integrations/linear", "integrations/notion", - "integrations/pi", "integrations/postgres", "integrations/raycast", "integrations/sentry", diff --git a/integrations/pi.mdx b/integrations/pi.mdx deleted file mode 100644 index df124ce..0000000 --- a/integrations/pi.mdx +++ /dev/null @@ -1,95 +0,0 @@ ---- -title: "Pi" -description: "A minimal, extensible terminal coding harness with multi-provider support." ---- - - - Pi - - -## Features - -- Multi-provider support: Anthropic, OpenAI, and OSS models (DeepSeek, Kimi, MiniMax, GLM) -- 15+ models across multiple providers -- Configurable reasoning levels: low, medium, high, max -- Minimal and extensible terminal-based workflow -- JSON output mode for structured results - -## Installation - - - - Install the `pi` CLI on your machine. See the [Pi documentation](https://github.com/anthropics/pi) for setup instructions. - - - Go to [Integrations](https://app.tembo.io/integrations) and enable Pi as a coding agent. - - - Pi configuration lives in `~/.pi/agent/`. Set your preferred model and reasoning level. - - - -## Usage - -### Select Pi as your agent - -When starting a session, specify Pi with a model and reasoning level using the agent key format: - -``` -pi:model:reasoningLevel -``` - -### Examples - -**Default model (Claude Sonnet):** - -``` -pi:claude-sonnet-4-6 -``` - -**With a reasoning level:** - -``` -pi:claude-sonnet-4-6:high -``` - -**Using a different provider:** - -``` -pi:gpt-5.4 -``` - -### Reasoning levels - -| Level | Description | -|-------|-------------| -| `low` | Fast responses, minimal deliberation | -| `medium` | Balanced speed and depth | -| `high` | Thorough analysis and planning | -| `max` | Maximum reasoning depth | - -## Agents - -See [Agents](/features/agents) for the full list of supported coding agents and harnesses. - -## Advanced - - - - Pi supports 15+ models across Anthropic, OpenAI, and OSS providers: - - **Anthropic:** Opus 4.7, Opus 4.6, Sonnet 4.6, Opus 4.5, Haiku 4.5 - - **OpenAI:** GPT-5.5, GPT-5.4, GPT-5.3 Codex, GPT-5.4 Mini, GPT-5.4 Nano, GPT-5.2 - - **OSS (via Fireworks AI):** GLM 5.1, Kimi K2.6, DeepSeek V4 Pro, MiniMax M2.7 - - - - Verify Pi is enabled on the [Integrations page](https://app.tembo.io/integrations). Ensure the `pi` CLI is installed and configured correctly on your machine. - - - - Update your Pi configuration in `~/.pi/agent/` to set a different default model, or specify the model directly in the agent key when starting a session. - -