link-assistant · konard · Apr 25, 2026 · Apr 23, 2026 · Apr 23, 2026 · Apr 23, 2026
diff --git a/.github/workflows/integration-tests.yml b/.github/workflows/integration-tests.yml
@@ -48,7 +48,7 @@ jobs:
           if [ -n "${{ inputs.test_pattern }}" ]; then
             bun test "${{ inputs.test_pattern }}"
           else
-            bun test
+            bun test ./tests/integration/*.js
           fi
         working-directory: js
 
@@ -98,7 +98,7 @@ jobs:
           if [ -n "${{ inputs.test_pattern }}" ]; then
             bun test "${{ inputs.test_pattern }}"
           else
-            bun test
+            bun test ./tests/integration/*.js
           fi
         working-directory: js
 

diff --git a/.github/workflows/js.yml b/.github/workflows/js.yml
@@ -134,7 +134,7 @@ jobs:
         working-directory: js
 
       - name: Run unit tests
-        run: bun test tests/json-standard-unit.test.js tests/process-name.test.js
+        run: bun test ./tests/json-standard-unit.js ./tests/process-name.js ./tests/cli.ts ./tests/cli_options.ts
         working-directory: js
 
       - name: Commit cached API responses
@@ -183,7 +183,7 @@ jobs:
         working-directory: js
 
       - name: Run verbose HTTP logging integration test
-        run: bun test tests/integration/verbose-hi.test.js
+        run: bun test ./tests/integration/verbose-hi.js
         working-directory: js
         timeout-minutes: 5
 

diff --git a/FREE_MODELS.md b/FREE_MODELS.md
@@ -9,29 +9,39 @@ This document lists all free AI models currently supported by the agent. Free mo
 Use any free model with the `--model` flag:
 
 ```bash
-echo "hello" | agent --model opencode/nemotron-3-super-free
+echo "hello" | agent --model opencode/minimax-m2.5-free
 ```
 
 ## OpenCode Zen Free Models
 
 [OpenCode Zen](https://opencode.ai/docs/zen/) offers curated, tested models. These free models require no authentication:
 
-| Model                   | Model ID                           | Context Window  | Description                                         |
-| ----------------------- | ---------------------------------- | --------------- | --------------------------------------------------- |
-| Nemotron 3 Super Free   | `opencode/nemotron-3-super-free`   | ~262,144        | **Default.** NVIDIA hybrid Mamba-Transformer, strong reasoning |
-| MiniMax M2.5 Free       | `opencode/minimax-m2.5-free`      | ~200,000        | Strong general-purpose performance                  |
-| GPT 5 Nano              | `opencode/gpt-5-nano`             | ~400,000        | Reliable OpenAI-powered free option                 |
-| Big Pickle              | `opencode/big-pickle`             | ~200,000        | Stealth model, free during evaluation period        |
+| Model                 | Model ID                         | Context Window | Description                                       |
+| --------------------- | -------------------------------- | -------------- | ------------------------------------------------- |
+| MiniMax M2.5 Free     | `opencode/minimax-m2.5-free`     | 204,800        | **Default.** Strong general-purpose performance   |
+| Ling 2.6 Flash Free   | `opencode/ling-2.6-flash-free`   | 262,100        | Fast free open-weight model                       |
+| Hy3 Preview Free      | `opencode/hy3-preview-free`      | 256,000        | Preview free model with reasoning support         |
+| Nemotron 3 Super Free | `opencode/nemotron-3-super-free` | 204,800        | NVIDIA free endpoint with strong reasoning        |
+| GPT 5 Nano            | `opencode/gpt-5-nano`            | 400,000        | OpenAI-powered free option and compaction default |
+| Big Pickle            | `opencode/big-pickle`            | 200,000        | Stealth model, free during evaluation period      |
+
+Source note: checked on April 23, 2026 against [OpenCode Zen](https://opencode.ai/docs/zen/), `https://opencode.ai/zen/v1/models`, and [models.dev](https://models.dev/api.json). The Zen models endpoint currently also lists `trinity-large-preview-free`, but models.dev marks it deprecated, so it is not recommended here.
 
 ### Usage Examples
 
 ```bash
-# Nemotron 3 Super Free (default)
-echo "hello" | agent --model opencode/nemotron-3-super-free
-
-# MiniMax M2.5 Free
+# MiniMax M2.5 Free (default)
 echo "hello" | agent --model opencode/minimax-m2.5-free
 
+# Ling 2.6 Flash Free
+echo "hello" | agent --model opencode/ling-2.6-flash-free
+
+# Hy3 Preview Free
+echo "hello" | agent --model opencode/hy3-preview-free
+
+# Nemotron 3 Super Free
+echo "hello" | agent --model opencode/nemotron-3-super-free
+
 # GPT 5 Nano
 echo "hello" | agent --model opencode/gpt-5-nano
 
@@ -45,14 +55,14 @@ echo "hello" | agent --model opencode/big-pickle
 
 [Kilo Gateway](https://kilo.ai/docs/gateway) provides access to 500+ AI models. These free models require no API key:
 
-| Model                 | Model ID                       | Context Window | Description                               |
-| --------------------- | ------------------------------ | -------------- | ----------------------------------------- |
-| GLM-5                 | `kilo/glm-5-free`              | 202,752 tokens | **Recommended.** Z.AI flagship model      |
-| GLM 4.5 Air           | `kilo/glm-4.5-air-free`        | 131,072 tokens | Free Z.AI model with agent capabilities   |
-| MiniMax M2.5          | `kilo/minimax-m2.5-free`       | 204,800 tokens | Strong general-purpose performance        |
-| DeepSeek R1           | `kilo/deepseek-r1-free`        | 163,840 tokens | Advanced reasoning model                  |
-| Giga Potato           | `kilo/giga-potato-free`        | 256,000 tokens | Free evaluation model                     |
-| Trinity Large Preview | `kilo/trinity-large-preview`   | 131,000 tokens | Arcee AI preview model                    |
+| Model                 | Model ID                     | Context Window | Description                             |
+| --------------------- | ---------------------------- | -------------- | --------------------------------------- |
+| GLM-5                 | `kilo/glm-5-free`            | 202,752 tokens | **Recommended.** Z.AI flagship model    |
+| GLM 4.5 Air           | `kilo/glm-4.5-air-free`      | 131,072 tokens | Free Z.AI model with agent capabilities |
+| MiniMax M2.5          | `kilo/minimax-m2.5-free`     | 204,800 tokens | Strong general-purpose performance      |
+| DeepSeek R1           | `kilo/deepseek-r1-free`      | 163,840 tokens | Advanced reasoning model                |
+| Giga Potato           | `kilo/giga-potato-free`      | 256,000 tokens | Free evaluation model                   |
+| Trinity Large Preview | `kilo/trinity-large-preview` | 131,000 tokens | Arcee AI preview model                  |
 
 ### Usage Examples
 
@@ -79,26 +89,28 @@ echo "hello" | agent --model kilo/giga-potato-free
 
 The following models were previously free but are no longer available:
 
-| Model              | Former Model ID               | Status                                   |
-| ------------------ | ----------------------------- | ---------------------------------------- |
-| Qwen 3.6 Plus Free | `opencode/qwen3.6-plus-free`  | Free promotion ended (April 2026) — now requires OpenCode Go subscription. See [issue #242](https://github.com/link-assistant/agent/issues/242) |
-| Kimi K2.5 Free     | `opencode/kimi-k2.5-free`     | Removed from OpenCode Zen (March 2026) — see [issue #208](https://github.com/link-assistant/agent/issues/208) |
-| Grok Code Fast 1   | `opencode/grok-code`          | Discontinued January 2026                |
-| MiniMax M2.1 Free  | `opencode/minimax-m2.1-free`  | Replaced by `opencode/minimax-m2.5-free` |
-| GLM 4.7 Free       | `opencode/glm-4.7-free`       | No longer free on OpenCode Zen           |
-| Kimi K2.5 (Kilo)   | `kilo/kimi-k2.5-free`         | Replaced by other Kilo free models       |
-| MiniMax M2.1 (Kilo)| `kilo/minimax-m2.1-free`      | Replaced by `kilo/minimax-m2.5-free`     |
+| Model               | Former Model ID              | Status                                                                                                                                          |
+| ------------------- | ---------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------- |
+| Qwen 3.6 Plus Free  | `opencode/qwen3.6-plus-free` | Free promotion ended (April 2026) — now requires OpenCode Go subscription. See [issue #242](https://github.com/link-assistant/agent/issues/242) |
+| Kimi K2.5 Free      | `opencode/kimi-k2.5-free`    | Removed from OpenCode Zen (March 2026) — see [issue #208](https://github.com/link-assistant/agent/issues/208)                                   |
+| Grok Code Fast 1    | `opencode/grok-code`         | Discontinued January 2026                                                                                                                       |
+| MiniMax M2.1 Free   | `opencode/minimax-m2.1-free` | Replaced by `opencode/minimax-m2.5-free`                                                                                                        |
+| GLM 4.7 Free        | `opencode/glm-4.7-free`      | No longer free on OpenCode Zen                                                                                                                  |
+| Kimi K2.5 (Kilo)    | `kilo/kimi-k2.5-free`        | Replaced by other Kilo free models                                                                                                              |
+| MiniMax M2.1 (Kilo) | `kilo/minimax-m2.1-free`     | Replaced by `kilo/minimax-m2.5-free`                                                                                                            |
 
 ---
 
 ## Choosing Between Providers
 
 ### Use OpenCode Zen when:
+
 - You want the most tested and reliable free models
-- You prefer `nemotron-3-super-free` as the default with ~262K context window
+- You prefer `minimax-m2.5-free` as the default with a 204,800 token context window
 - You need a simple, curated list of models
 
 ### Use Kilo Gateway when:
+
 - You want access to GLM-5 (currently free, limited time)
 - You need larger context windows (up to 256,000 tokens)
 - You want more free model options
@@ -108,6 +120,9 @@ The following models were previously free but are no longer available:
 The agent intelligently routes model requests:
 
 - `nemotron-3-super-free` without provider prefix → OpenCode Zen (`opencode/nemotron-3-super-free`)
+- `minimax-m2.5-free` without provider prefix → OpenCode Zen (`opencode/minimax-m2.5-free`)
+- `ling-2.6-flash-free` without provider prefix → OpenCode Zen (`opencode/ling-2.6-flash-free`)
+- `hy3-preview-free` without provider prefix → OpenCode Zen (`opencode/hy3-preview-free`)
 - `big-pickle` without provider prefix → OpenCode Zen (`opencode/big-pickle`)
 - `kilo/minimax-m2.5-free` explicitly → Kilo Gateway