Skip to content

A self-improving RLM agent for coding workflows and long-running autonomous tasks.

Rust

show model and context usage in the prompt bar (#252)

ktc17c926dcb42
diff --git a/packages/ai/scripts/generate-models.ts b/packages/ai/scripts/generate-models.ts
index 34ae7b1c64..443220fb98 100644
--- a/packages/ai/scripts/generate-models.ts
+++ b/packages/ai/scripts/generate-models.ts
@@ -430,7 +430,7 @@ function createPrimeInferenceModel(
): Model<"openai-completions"> {
return {
id: entry.id,
- name: `${getPrimeInferenceDisplayName(entry.id)} (Prime Inference)`,
+ name: getPrimeInferenceDisplayName(entry.id),
api: "openai-completions",
provider: "prime-inference",
baseUrl: PRIME_INFERENCE_BASE_URL,
diff --git a/packages/ai/src/models.generated.ts b/packages/ai/src/models.generated.ts
index 049272d94b..78ca58e603 100644
--- a/packages/ai/src/models.generated.ts
+++ b/packages/ai/src/models.generated.ts
@@ -12800,7 +12800,7 @@ export const MODELS = {
"prime-inference": {
"Qwen/Qwen3-235B-A22B-Thinking-2507": {
id: "Qwen/Qwen3-235B-A22B-Thinking-2507",
- name: "QWEN3 235B A22B Thinking 2507 (Prime Inference)",
+ name: "QWEN3 235B A22B Thinking 2507",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
@@ -12818,7 +12818,7 @@ export const MODELS = {
} satisfies Model<"openai-completions">,
"anthropic/claude-haiku-4.5": {
id: "anthropic/claude-haiku-4.5",
- name: "Claude Haiku 4.5 (Prime Inference)",
+ name: "Claude Haiku 4.5",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
@@ -12836,7 +12836,7 @@ export const MODELS = {
} satisfies Model<"openai-completions">,
"anthropic/claude-opus-4.6": {
id: "anthropic/claude-opus-4.6",
- name: "Claude Opus 4.6 (Prime Inference)",
+ name: "Claude Opus 4.6",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
@@ -12855,7 +12855,7 @@ export const MODELS = {
} satisfies Model<"openai-completions">,
"anthropic/claude-opus-4.7": {
id: "anthropic/claude-opus-4.7",
- name: "Claude Opus 4.7 (Prime Inference)",
+ name: "Claude Opus 4.7",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
@@ -12874,7 +12874,7 @@ export const MODELS = {
} satisfies Model<"openai-completions">,
"anthropic/claude-opus-4.8": {
id: "anthropic/claude-opus-4.8",
- name: "Claude Opus 4.8 (Prime Inference)",
+ name: "Claude Opus 4.8",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
@@ -12893,7 +12893,7 @@ export const MODELS = {
} satisfies Model<"openai-completions">,
"anthropic/claude-sonnet-4.5": {
id: "anthropic/claude-sonnet-4.5",
- name: "Claude Sonnet 4.5 (Prime Inference)",
+ name: "Claude Sonnet 4.5",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
@@ -12911,7 +12911,7 @@ export const MODELS = {
} satisfies Model<"openai-completions">,
"anthropic/claude-sonnet-4.6": {
id: "anthropic/claude-sonnet-4.6",
- name: "Claude Sonnet 4.6 (Prime Inference)",
+ name: "Claude Sonnet 4.6",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
@@ -12930,7 +12930,7 @@ export const MODELS = {
} satisfies Model<"openai-completions">,
"deepseek/deepseek-v3.2": {
id: "deepseek/deepseek-v3.2",
- name: "Deepseek V3.2 (Prime Inference)",
+ name: "Deepseek V3.2",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
@@ -12948,7 +12948,7 @@ export const MODELS = {
} satisfies Model<"openai-completions">,
"deepseek/deepseek-v4-flash": {
id: "deepseek/deepseek-v4-flash",
- name: "Deepseek V4 Flash (Prime Inference)",
+ name: "Deepseek V4 Flash",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
@@ -12967,7 +12967,7 @@ export const MODELS = {
} satisfies Model<"openai-completions">,
"deepseek/deepseek-v4-pro": {
id: "deepseek/deepseek-v4-pro",
- name: "Deepseek V4 PRO (Prime Inference)",
+ name: "Deepseek V4 PRO",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
@@ -12986,7 +12986,7 @@ export const MODELS = {
} satisfies Model<"openai-completions">,
"minimax/minimax-m3": {
id: "minimax/minimax-m3",
- name: "Minimax M3 (Prime Inference)",
+ name: "Minimax M3",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
@@ -13004,7 +13004,7 @@ export const MODELS = {
} satisfies Model<"openai-completions">,
"moonshotai/kimi-k2.7-code": {
id: "moonshotai/kimi-k2.7-code",
- name: "Kimi K2.7 Code (Prime Inference)",
+ name: "Kimi K2.7 Code",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
@@ -13022,7 +13022,7 @@ export const MODELS = {
} satisfies Model<"openai-completions">,
"nvidia/nemotron-3-nano-30b-a3b": {
id: "nvidia/nemotron-3-nano-30b-a3b",
- name: "Nemotron 3 Nano 30B A3B (Prime Inference)",
+ name: "Nemotron 3 Nano 30B A3B",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
@@ -13040,7 +13040,7 @@ export const MODELS = {
} satisfies Model<"openai-completions">,
"nvidia/nemotron-3-super-120b-a12b": {
id: "nvidia/nemotron-3-super-120b-a12b",
- name: "Nemotron 3 Super 120B A12B (Prime Inference)",
+ name: "Nemotron 3 Super 120B A12B",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
@@ -13058,7 +13058,7 @@ export const MODELS = {
} satisfies Model<"openai-completions">,
"openai/gpt-5.3-codex": {
id: "openai/gpt-5.3-codex",
- name: "GPT 5.3 Codex (Prime Inference)",
+ name: "GPT 5.3 Codex",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
@@ -13077,7 +13077,7 @@ export const MODELS = {
} satisfies Model<"openai-completions">,
"openai/gpt-5.4": {
id: "openai/gpt-5.4",
- name: "GPT 5.4 (Prime Inference)",
+ name: "GPT 5.4",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
@@ -13096,7 +13096,7 @@ export const MODELS = {
} satisfies Model<"openai-completions">,
"openai/gpt-5.4-mini": {
id: "openai/gpt-5.4-mini",
- name: "GPT 5.4 Mini (Prime Inference)",
+ name: "GPT 5.4 Mini",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
@@ -13115,7 +13115,7 @@ export const MODELS = {
} satisfies Model<"openai-completions">,
"openai/gpt-5.4-pro": {
id: "openai/gpt-5.4-pro",
- name: "GPT 5.4 PRO (Prime Inference)",
+ name: "GPT 5.4 PRO",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
@@ -13134,7 +13134,7 @@ export const MODELS = {
} satisfies Model<"openai-completions">,
"openai/gpt-5.5": {
id: "openai/gpt-5.5",
- name: "GPT 5.5 (Prime Inference)",
+ name: "GPT 5.5",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
@@ -13153,7 +13153,7 @@ export const MODELS = {
} satisfies Model<"openai-completions">,
"prime-intellect/intellect-3": {
id: "prime-intellect/intellect-3",
- name: "Intellect 3 (Prime Inference)",
+ name: "Intellect 3",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
@@ -13171,7 +13171,7 @@ export const MODELS = {
} satisfies Model<"openai-completions">,
"qwen/qwen3-coder-next": {
id: "qwen/qwen3-coder-next",
- name: "QWEN3 Coder Next (Prime Inference)",
+ name: "QWEN3 Coder Next",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
@@ -13189,7 +13189,7 @@ export const MODELS = {
} satisfies Model<"openai-completions">,
"qwen/qwen3-max": {
id: "qwen/qwen3-max",
- name: "QWEN3 MAX (Prime Inference)",
+ name: "QWEN3 MAX",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
@@ -13207,7 +13207,7 @@ export const MODELS = {
} satisfies Model<"openai-completions">,
"qwen/qwen3-vl-235b-a22b-thinking": {
id: "qwen/qwen3-vl-235b-a22b-thinking",
- name: "QWEN3 VL 235B A22B Thinking (Prime Inference)",
+ name: "QWEN3 VL 235B A22B Thinking",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
@@ -13225,7 +13225,7 @@ export const MODELS = {
} satisfies Model<"openai-completions">,
"x-ai/grok-4.20": {
id: "x-ai/grok-4.20",
- name: "Grok 4.20 (Prime Inference)",
+ name: "Grok 4.20",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
@@ -13243,7 +13243,7 @@ export const MODELS = {
} satisfies Model<"openai-completions">,
"x-ai/grok-4.20-multi-agent": {
id: "x-ai/grok-4.20-multi-agent",
- name: "Grok 4.20 Multi Agent (Prime Inference)",
+ name: "Grok 4.20 Multi Agent",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
@@ -13261,7 +13261,7 @@ export const MODELS = {
} satisfies Model<"openai-completions">,
"z-ai/glm-5": {
id: "z-ai/glm-5",
- name: "GLM 5 (Prime Inference)",
+ name: "GLM 5",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
@@ -13279,7 +13279,7 @@ export const MODELS = {
} satisfies Model<"openai-completions">,
"z-ai/glm-5.1": {
id: "z-ai/glm-5.1",
- name: "GLM 5.1 (Prime Inference)",
+ name: "GLM 5.1",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
@@ -13297,7 +13297,7 @@ export const MODELS = {
} satisfies Model<"openai-completions">,
"z-ai/glm-5.2": {
id: "z-ai/glm-5.2",
- name: "GLM 5.2 (Prime Inference)",
+ name: "GLM 5.2",
api: "openai-completions",
provider: "prime-inference",
baseUrl: "https://api.pinference.ai/api/v1",
diff --git a/packages/coding-agent/src/modes/interactive/interactive-mode.ts b/packages/coding-agent/src/modes/interactive/interactive-mode.ts
index 07a32ec1e1..084c595520 100644
--- a/packages/coding-agent/src/modes/interactive/interactive-mode.ts
+++ b/packages/coding-agent/src/modes/interactive/interactive-mode.ts
@@ -4427,9 +4427,20 @@ export class InteractiveMode {
}
private getTrayLocationLabel(): string | undefined {
- const location = this.footerDataProvider.getGitBranch() ?? formatSplashCwd(this.getCurrentCwd());
const agentsHint = this.getAgentsViewTrayHint();
- return [agentsHint, location].filter((label): label is string => label !== undefined).join(" ");
+ return [agentsHint, this.getModelTrayLabel()].filter((label): label is string => label !== undefined).join(" ");
+ }
+
+ private getModelTrayLabel(): string {
+ const model = this.getCurrentModel();
+ if (!model) {
+ return "—";
+ }
+ if (!model.reasoning) {
+ return model.name;
+ }
+ const level = this.connectionState?.thinkingLevel ?? "off";
+ return level === "off" ? model.name : `${model.name} • ${level}`;
}
private getAgentsViewTrayHint(): string | undefined {
@@ -4445,14 +4456,10 @@ export class InteractiveMode {
private getTrayContextLabel(): string | undefined {
const goalLabel = this.getTrayGoalLabel();
const usage = this.getConnectionContextUsage();
- if (!usage) {
- return goalLabel;
- }
- if (usage.percent === null) {
- return goalLabel;
- }
- const remainingPercent = Math.max(0, Math.min(100, 100 - usage.percent));
- const contextLabel = remainingPercent <= 50 ? `${Math.round(remainingPercent)}% context left` : undefined;
+ const contextLabel =
+ usage && typeof usage.tokens === "number" && typeof usage.percent === "number"
+ ? `${formatTokenCount(usage.tokens)} (${Math.round(usage.percent)}%)`
+ : undefined;
return [goalLabel, contextLabel].filter((label) => label !== undefined).join(" · ") || undefined;
}
diff --git a/packages/coding-agent/test/interactive-mode-status.test.ts b/packages/coding-agent/test/interactive-mode-status.test.ts
index b719d63b88..9df35ee047 100644
--- a/packages/coding-agent/test/interactive-mode-status.test.ts
+++ b/packages/coding-agent/test/interactive-mode-status.test.ts
@@ -1155,12 +1155,13 @@ describe("InteractiveMode goal status announcements", () => {
});
describe("InteractiveMode tray goal label", () => {
+ type TrayUsage = { contextWindow: number; tokens: number | null; percent: number | null };
type TrayLabelHarness = {
connectionState: {
goal: GoalState;
- contextUsage: { contextWindow: number; percent: number | null } | undefined;
+ contextUsage: TrayUsage | undefined;
};
- uiServices: { getContextUsage(): { contextWindow: number; percent: number | null } | undefined };
+ uiServices: { getContextUsage(): TrayUsage | undefined };
getTrayContextLabel(): string | undefined;
};
const getTrayContextLabel = (InteractiveMode.prototype as unknown as TrayLabelHarness).getTrayContextLabel;
@@ -1183,7 +1184,7 @@ describe("InteractiveMode tray goal label", () => {
expect(getTrayContextLabel.call(fakeThis)).toBe("Pursuing goal (1m 05s)");
});
- test("combines active goals with low-context signal in one lower-tray label", () => {
+ test("combines active goals with token/context usage in one lower-tray label", () => {
const fakeThis = Object.create(InteractiveMode.prototype) as TrayLabelHarness;
fakeThis.connectionState = {
goal: {
@@ -1194,11 +1195,29 @@ describe("InteractiveMode tray goal label", () => {
timeUsedSeconds: 65,
continuationsUsed: 1,
} satisfies GoalState,
- contextUsage: { contextWindow: 100_000, percent: 75 },
+ contextUsage: { contextWindow: 100_000, tokens: 75_000, percent: 75 },
};
fakeThis.uiServices = { getContextUsage: () => undefined };
- expect(getTrayContextLabel.call(fakeThis)).toBe("Pursuing goal (1m 05s) · 25% context left");
+ expect(getTrayContextLabel.call(fakeThis)).toBe("Pursuing goal (1m 05s) · 75k (75%)");
+ });
+
+ test("omits the usage segment when token count is unknown", () => {
+ const fakeThis = Object.create(InteractiveMode.prototype) as TrayLabelHarness;
+ fakeThis.connectionState = {
+ goal: {
+ active: true,
+ status: "active",
+ objective: "finish the task",
+ tokensUsed: 0,
+ timeUsedSeconds: 65,
+ continuationsUsed: 1,
+ } satisfies GoalState,
+ contextUsage: { contextWindow: 100_000, tokens: null, percent: null },
+ };
+ fakeThis.uiServices = { getContextUsage: () => undefined };
+
+ expect(getTrayContextLabel.call(fakeThis)).toBe("Pursuing goal (1m 05s)");
});
});