Skip to content

Refresh MiniMax M3 model params #1374

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
omeraplak merged 1 commit into from
Jul 8, 2026
Merged

Refresh MiniMax M3 model params #1374

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions packages/core/src/registries/model-provider-registry-minimax.spec.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -149,6 +149,7 @@ describe("MiniMax provider registry", () => {
const registry = ModelProviderRegistry.getInstance();

const modelIds = [
"MiniMax-M3",
"MiniMax-M2.7",
"MiniMax-M2.7-highspeed",
"MiniMax-M2.5",
Expand Down
4 changes: 2 additions & 2 deletions packages/core/src/registries/model-provider-registry.generated.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -317,15 +317,15 @@ export const MODEL_PROVIDER_REGISTRY: Record<string, ModelProviderRegistryEntry>
npm: "@ai-sdk/anthropic",
api: "https://api.minimax.io/anthropic/v1",
env: ["MINIMAX_API_KEY"],
doc: "https://platform.minimax.io/docs/guides/quickstart",
doc: "https://platform.minimax.io/docs/api-reference/api-overview",
},
"minimax-cn": {
id: "minimax-cn",
name: "MiniMax (China)",
npm: "@ai-sdk/anthropic",
api: "https://api.minimaxi.com/anthropic/v1",
env: ["MINIMAX_API_KEY"],
doc: "https://platform.minimaxi.com/docs/guides/quickstart",
doc: "https://platform.minimaxi.com/docs/api-reference/api-overview",
Comment on lines 322 to +328

@coderabbitai coderabbitai Bot Jul 8, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🗄️ Data Integrity & Integration | 🟠 Major | ⚡ Quick win

Align the generated MiniMax China entry with the runtime registry.

model-provider-registry.generated.ts still routes minimax-cn through @ai-sdk/anthropic and /anthropic/v1, but model-provider-registry.ts now uses @ai-sdk/openai-compatible and /v1. That leaves the generated metadata and runtime registry out of sync, so consumers can resolve the same provider through different transports. Please regenerate or update this entry so both files agree.

Suggested alignment 
-    npm: "`@ai-sdk/anthropic`",
-    api: "https://api.minimaxi.com/anthropic/v1",
+    npm: "`@ai-sdk/openai-compatible`",
+    api: "https://api.minimaxi.com/v1",
📝 Committable suggestion

‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.

Suggested change
"minimax-cn": {
id: "minimax-cn",
name: "MiniMax (China)",
npm: "@ai-sdk/anthropic",
api: "https://api.minimaxi.com/anthropic/v1",
env: ["MINIMAX_API_KEY"],
doc: "https://platform.minimaxi.com/docs/guides/quickstart",
doc: "https://platform.minimaxi.com/docs/api-reference/api-overview",
"minimax-cn": {
id: "minimax-cn",
name: "MiniMax (China)",
npm: "`@ai-sdk/openai-compatible`",
api: "https://api.minimaxi.com/v1",
env: ["MINIMAX_API_KEY"],
doc: "https://platform.minimaxi.com/docs/api-reference/api-overview",
🤖 Prompt for AI Agents 
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@packages/core/src/registries/model-provider-registry.generated.ts` around
lines 322 - 328, The generated MiniMax China provider metadata is out of sync
with the runtime registry. Update the `minimax-cn` entry in
`model-provider-registry.generated.ts` to match `model-provider-registry.ts`,
including the transport package and base URL used by the `minimax-cn` provider
definition. Keep the `id`, `name`, `env`, and other shared fields aligned, and
ensure the generated registry reflects the same `@ai-sdk/openai-compatible` and
`/v1` configuration as the runtime source.

},
mistral: {
id: "mistral",
Expand Down
4 changes: 2 additions & 2 deletions packages/core/src/registries/model-provider-registry.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -150,15 +150,15 @@ const EXTRA_PROVIDER_REGISTRY: ModelProviderRegistryEntry[] = [
npm: "@ai-sdk/openai-compatible",
api: "https://api.minimax.io/v1",
env: ["MINIMAX_API_KEY"],
doc: "https://platform.minimax.io/docs/guides/quickstart",
doc: "https://platform.minimax.io/docs/api-reference/api-overview",
},
{
id: "minimax-cn",
name: "MiniMax (China)",
npm: "@ai-sdk/openai-compatible",
api: "https://api.minimaxi.com/v1",
env: ["MINIMAX_API_KEY"],
doc: "https://platform.minimaxi.com/docs/guides/quickstart",
doc: "https://platform.minimaxi.com/docs/api-reference/api-overview",
},
];

Expand Down
4 changes: 2 additions & 2 deletions packages/core/src/registries/model-provider-types.generated.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1103,8 +1103,8 @@ export type ProviderModelsMap = {
"qwen/qwen3-coder-30b",
];
readonly lucidquery: readonly ["lucidnova-rf1-100b", "lucidquery-nexus-coder"];
readonly minimax: readonly ["MiniMax-M2", "MiniMax-M2.1"];
readonly "minimax-cn": readonly ["MiniMax-M2", "MiniMax-M2.1"];
readonly minimax: readonly ["MiniMax-M2", "MiniMax-M2.1", "MiniMax-M3"];
readonly "minimax-cn": readonly ["MiniMax-M2", "MiniMax-M2.1", "MiniMax-M3"];
readonly mistral: readonly [
"codestral-latest",
"devstral-2512",
Expand Down
13 changes: 10 additions & 3 deletions website/models-docs/providers/minimax-cn.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,7 +14,7 @@ import { Agent } from "@voltagent/core";
const agent = new Agent({
name: "minimax-cn-agent",
instructions: "You are a helpful assistant",
model: "minimax-cn/MiniMax-M2.7",
model: "minimax-cn/MiniMax-M3",
});
```

Expand All @@ -34,13 +34,20 @@ You can override the base URL by setting `MINIMAX_CN_BASE_URL`.

## Provider docs

- https://platform.minimaxi.com/docs/guides/quickstart
- https://platform.minimaxi.com/docs/api-reference/api-overview

## Pricing

| Model | Input | Output | Cache read | Cache write |
|---|---:|---:|---:|---:|
| MiniMax-M3 | $0.60 / 1M tokens | $2.40 / 1M tokens | $0.12 / 1M tokens | Not listed |
Comment on lines +39 to +43

@coderabbitai coderabbitai Bot Jul 8, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🎯 Functional Correctness | 🟠 Major | ⚡ Quick win

Document the full M3 pricing split.

The current MiniMax pay-as-you-go page lists separate ≤512K and 512K~1M M3 pricing tiers, but this table only shows one rate set, so long-context usage will be underquoted. (platform.minimax.io)

🤖 Prompt for AI Agents 
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@website/models-docs/providers/minimax-cn.md` around lines 39 - 43, The
MiniMax-M3 pricing table is incomplete and only reflects one rate set, so update
the pricing details in the MiniMax provider docs to include the full split for
the ≤512K and 512K~1M tiers. Use the existing pricing section in the MiniMax
docs and the MiniMax-M3 row to add the missing tiered input/output/cache rates
so long-context costs are represented accurately.


## Models

| Model | Context | Description |
|---|---|---|
| MiniMax-M2.7 | 1M tokens | Latest flagship model |
| MiniMax-M3 | 1M tokens | Latest flagship model |
| MiniMax-M2.7 | 1M tokens | Previous generation |
| MiniMax-M2.7-highspeed | 1M tokens | Optimized for speed |
| MiniMax-M2.5 | 1M tokens | Previous generation |
| MiniMax-M2.5-highspeed | 204K tokens | Fast inference |
Expand Down
13 changes: 10 additions & 3 deletions website/models-docs/providers/minimax.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,7 +16,7 @@ import { Agent } from "@voltagent/core";
const agent = new Agent({
name: "minimax-agent",
instructions: "You are a helpful assistant",
model: "minimax/MiniMax-M2.7",
model: "minimax/MiniMax-M3",
});
```

Expand All @@ -36,13 +36,20 @@ You can override the base URL by setting `MINIMAX_BASE_URL`.

## Provider docs

- https://platform.minimax.io/docs/guides/quickstart
- https://platform.minimax.io/docs/api-reference/api-overview

## Pricing

| Model | Input | Output | Cache read | Cache write |
|---|---:|---:|---:|---:|
| MiniMax-M3 | $0.60 / 1M tokens | $2.40 / 1M tokens | $0.12 / 1M tokens | Not listed |
Comment on lines +41 to +45

@coderabbitai coderabbitai Bot Jul 8, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🎯 Functional Correctness | 🟠 Major | ⚡ Quick win

Document the full M3 pricing split.

The current MiniMax pay-as-you-go page lists separate ≤512K and 512K~1M M3 pricing tiers, but this table only shows one rate set, so long-context usage will be underquoted. (platform.minimax.io)

🤖 Prompt for AI Agents 
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@website/models-docs/providers/minimax.md` around lines 41 - 45, The
MiniMax-M3 pricing table is incomplete because it only lists a single rate set
and omits the separate ≤512K and 512K~1M tiers, so update the pricing section in
minimax.md to reflect the full M3 split shown on the MiniMax pay-as-you-go page.
Keep the existing MiniMax-M3 row structure, but add the distinct tiered pricing
entries with the correct input/output/cache values so long-context usage is
represented accurately.


## Models

| Model | Context | Description |
|---|---|---|
| MiniMax-M2.7 | 1M tokens | Latest flagship model |
| MiniMax-M3 | 1M tokens | Latest flagship model |
| MiniMax-M2.7 | 1M tokens | Previous generation |
| MiniMax-M2.7-highspeed | 1M tokens | Optimized for speed |
| MiniMax-M2.5 | 1M tokens | Previous generation |
| MiniMax-M2.5-highspeed | 204K tokens | Fast inference |
Expand Down
Loading