ceos/unsloth
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

studio/chat: drive ChatSettingsPanel from a per-provider capability map

Browse source 
Replace the binary isExternalModel toggle in the sampling section with a
provider-aware capability map. Each external provider type advertises
which of top_k / min_p / repetition_penalty / presence_penalty its
chat-completions API actually accepts, so the panel only renders the
knobs that map onto the active provider's request body.

Anthropic now exposes top_k; DeepSeek hides presence_penalty (deprecated
in their docs); OpenRouter and custom providers continue to show every
knob (OpenRouter drops unsupported server-side, custom assumes
OpenAI-compat or a permissive vLLM/Ollama backend). Local models are
unaffected — null capabilities means 'show everything'.

chat-adapter.ts now forwards top_k / presence_penalty to the external
proxy only when the active provider's capabilities permit it, so the
request body matches what the UI shows.
This commit is contained in:
Roland Tannous 2026-05-12 09:22:53 +04:00
commit f40b5fbe75
4 changed files with 176 additions and 46 deletions
Download patch file Download diff file Expand all files Collapse all files

13
studio/frontend/src/features/chat/api/chat-adapter.ts
View file

@ -26,6 +26,7 @@ import {
loadExternalProviders,
parseExternalModelId,
} from "../external-providers";
import { getProviderCapabilities } from "../provider-capabilities";
import { useChatRuntimeStore } from "../stores/chat-runtime-store";
import { isMultimodalResponse } from "../types/api";
import type { ChatModelSummary } from "../types/runtime";
@ -847,6 +848,9 @@ export function createOpenAIStreamAdapter(): ChatModelAdapter {
externalProvider?.providerType === "custom"
? "openai"
: externalProvider?.providerType;
const externalCapabilities = getProviderCapabilities(
externalProvider?.providerType,
);
const buildRequestPayload = async (forceRefreshPublicKey = false) => {
if (externalSelection && externalProvider) {
return {
@ -856,7 +860,14 @@ export function createOpenAIStreamAdapter(): ChatModelAdapter {
temperature: params.temperature,
top_p: params.topP,
max_tokens: params.maxTokens,
presence_penalty: params.presencePenalty,
// Only forward sampling knobs the provider actually accepts; the
// backend's external-provider proxy is param-permissive and would
// surface a 400 from providers that reject unknown fields (e.g.
// OpenAI rejects top_k, Anthropic/DeepSeek reject presence_penalty).
...(externalCapabilities?.topK ? { top_k: params.topK } : {}),
...(externalCapabilities?.presencePenalty
? { presence_penalty: params.presencePenalty }
: {}),
provider_id: externalProvider.id,
provider_type: externalBackendProviderType,
external_model: externalSelection.modelId,

16
studio/frontend/src/features/chat/chat-page.tsx
View file

@ -42,7 +42,12 @@ import { ChatSettingsPanel } from "./chat-settings-sheet";
import { ContextUsageBar } from "./components/context-usage-bar";
import { ModelLoadInlineStatus } from "./components/model-load-status";
import { db } from "./db";
import { buildExternalModelId, isExternalModelId } from "./external-providers";
import {
buildExternalModelId,
isExternalModelId,
parseExternalModelId,
} from "./external-providers";
import { getProviderCapabilities } from "./provider-capabilities";
import { useChatModelRuntime } from "./hooks/use-chat-model-runtime";
import {
clearTrainingCompareHandoff,
@ -616,6 +621,14 @@ export function ChatPage(): ReactElement {
() => isExternalModelId(inferenceParams.checkpoint),
[inferenceParams.checkpoint],
);
const activeProviderCapabilities = useMemo(() => {
const selection = parseExternalModelId(inferenceParams.checkpoint);
if (!selection) return null;
const provider = externalProviders.find(
(p) => p.id === selection.providerId,
);
return getProviderCapabilities(provider?.providerType);
}, [externalProviders, inferenceParams.checkpoint]);
const canCompare = useMemo(() => {
return Boolean(inferenceParams.checkpoint) && !isExternalModel;
}, [inferenceParams.checkpoint, isExternalModel]);
@ -1188,6 +1201,7 @@ export function ChatPage(): ReactElement {
params={inferenceParams}
onParamsChange={setInferenceParams}
isExternalModel={isExternalModel}
providerCapabilities={activeProviderCapabilities}
onReloadModel={() => {
const state = useChatRuntimeStore.getState();
if (state.params.checkpoint) {

110
studio/frontend/src/features/chat/chat-settings-sheet.tsx
View file

@ -80,6 +80,7 @@ import {
toPresetParams,
type Preset,
} from "./presets/preset-policy";
import type { ProviderCapabilities } from "./provider-capabilities";
import type { InferenceParams } from "./types/runtime";
export { defaultInferenceParams, type Preset } from "./presets/preset-policy";
@ -506,6 +507,12 @@ interface ChatSettingsPanelProps {
params: InferenceParams;
onParamsChange: (params: InferenceParams) => void;
isExternalModel?: boolean;
/**
* Sampling-param capability set for the active external provider, or `null`
* for local models (in which case every knob is rendered). Drives the
* per-param visibility in the sampling section.
*/
providerCapabilities?: ProviderCapabilities | null;
onReloadModel?: () => void;
}
@ -515,8 +522,19 @@ export function ChatSettingsPanel({
params,
onParamsChange,
isExternalModel = false,
providerCapabilities = null,
onReloadModel,
}: ChatSettingsPanelProps) {
// For non-external (local) models we show every knob — providerCapabilities
// is only consulted when `isExternalModel` is true. An external model with an
// unknown provider falls back to the OpenAI-compat shape via
// getProviderCapabilities, so these flags never undercount support.
const showTopK = !isExternalModel || Boolean(providerCapabilities?.topK);
const showMinP = !isExternalModel || Boolean(providerCapabilities?.minP);
const showRepetitionPenalty =
!isExternalModel || Boolean(providerCapabilities?.repetitionPenalty);
const showPresencePenalty =
!isExternalModel || Boolean(providerCapabilities?.presencePenalty);
const isMobile = useIsMobile();
const isGguf = useChatRuntimeStore((s) => s.activeGgufVariant) != null;
const hasModelContent =
@ -1153,51 +1171,55 @@ export function ChatSettingsPanel({
displayValue={params.topP === 1 ? "Off" : undefined}
info="Nucleus sampling. Restricts choices to the smallest set of tokens whose cumulative probability reaches this threshold. 1.0 = off."
/>
{!isExternalModel ? (
<>
<ParamSlider
label="Top K"
value={params.topK}
min={0}
max={100}
step={1}
onChange={set("topK")}
displayValue={params.topK === 0 ? "Off" : undefined}
info="Limits sampling to the K most likely tokens at each step. 0 = off."
/>
<ParamSlider
label="Min P"
value={params.minP}
min={0}
max={1}
step={0.01}
onChange={set("minP")}
info="Drops tokens whose probability is below this fraction of the top token's probability. Filters unlikely candidates."
/>
<ParamSlider
label="Repetition Penalty"
value={params.repetitionPenalty}
min={1}
max={2}
step={0.05}
onChange={set("repetitionPenalty")}
displayValue={
params.repetitionPenalty === 1 ? "Off" : undefined
}
info="Down-weights tokens that have already appeared, reducing repetition. 1.0 = off; higher values penalize more strongly."
/>
</>
{showTopK ? (
<ParamSlider
label="Top K"
value={params.topK}
min={0}
max={100}
step={1}
onChange={set("topK")}
displayValue={params.topK === 0 ? "Off" : undefined}
info="Limits sampling to the K most likely tokens at each step. 0 = off."
/>
) : null}
{showMinP ? (
<ParamSlider
label="Min P"
value={params.minP}
min={0}
max={1}
step={0.01}
onChange={set("minP")}
info="Drops tokens whose probability is below this fraction of the top token's probability. Filters unlikely candidates."
/>
) : null}
{showRepetitionPenalty ? (
<ParamSlider
label="Repetition Penalty"
value={params.repetitionPenalty}
min={1}
max={2}
step={0.05}
onChange={set("repetitionPenalty")}
displayValue={
params.repetitionPenalty === 1 ? "Off" : undefined
}
info="Down-weights tokens that have already appeared, reducing repetition. 1.0 = off; higher values penalize more strongly."
/>
) : null}
{showPresencePenalty ? (
<ParamSlider
label="Presence Penalty"
value={params.presencePenalty}
min={0}
max={2}
step={0.1}
onChange={set("presencePenalty")}
displayValue={params.presencePenalty === 0 ? "Off" : undefined}
info="Penalizes any token that has already appeared at least once, encouraging the model to introduce new topics. 0 = off."
/>
) : null}
<ParamSlider
label="Presence Penalty"
value={params.presencePenalty}
min={0}
max={2}
step={0.1}
onChange={set("presencePenalty")}
displayValue={params.presencePenalty === 0 ? "Off" : undefined}
info="Penalizes any token that has already appeared at least once, encouraging the model to introduce new topics. 0 = off."
/>
{!isExternalModel && !isGguf && (
<ParamSlider
label="Max Seq Length"

83
studio/frontend/src/features/chat/provider-capabilities.ts Normal file
View file

@ -0,0 +1,83 @@
// SPDX-License-Identifier: AGPL-3.0-only
// Copyright 2026-present the Unsloth AI Inc. team. All rights reserved. See /studio/LICENSE.AGPL-3.0
/**
* Per-provider sampling parameter capability matrix.
*
* Values are derived from each provider's published chat-completion docs as of
* 2026-05. They describe which of our UI knobs map cleanly onto the provider's
* request body; the panel hides params a provider does not accept so users
* cannot dial a value that gets silently dropped or rejected.
*
* "Local" models (anything that is not an external provider) are represented by
* a null capability — every knob renders for them.
*/
export interface ProviderCapabilities {
/** top-k token sampling (only Anthropic on the providers we ship). */
topK: boolean;
/** min-p token cutoff (no SaaS provider currently exposes this). */
minP: boolean;
/** Repetition penalty (no SaaS provider currently exposes this). */
repetitionPenalty: boolean;
/** OpenAI-style presence penalty. */
presencePenalty: boolean;
}
const OPENAI_COMPAT_BASE: ProviderCapabilities = {
topK: false,
minP: false,
repetitionPenalty: false,
presencePenalty: true,
};
const ALL_SUPPORTED: ProviderCapabilities = {
topK: true,
minP: true,
repetitionPenalty: true,
presencePenalty: true,
};
const PROVIDER_CAPABILITIES: Record<string, ProviderCapabilities> = {
openai: OPENAI_COMPAT_BASE,
// Anthropic's Messages API accepts top_k but not presence/frequency penalty.
anthropic: {
topK: true,
minP: false,
repetitionPenalty: false,
presencePenalty: false,
},
mistral: OPENAI_COMPAT_BASE,
gemini: OPENAI_COMPAT_BASE,
// DeepSeek deprecated presence/frequency penalty in their current docs.
deepseek: {
topK: false,
minP: false,
repetitionPenalty: false,
presencePenalty: false,
},
kimi: OPENAI_COMPAT_BASE,
qwen: OPENAI_COMPAT_BASE,
huggingface: OPENAI_COMPAT_BASE,
// OpenRouter silently drops params the target model does not support, so we
// surface every knob and let the gateway handle the per-model fan-out.
openrouter: ALL_SUPPORTED,
// Custom providers are assumed OpenAI-compatible by the backend; users who
// point at vLLM/Ollama backends often want top_k / min_p / repetition,
// so be permissive.
custom: ALL_SUPPORTED,
};
const DEFAULT_EXTERNAL_CAPABILITIES = OPENAI_COMPAT_BASE;
/**
* Resolve the capability set for an external provider. Returns `null` for
* a local model (i.e. when `providerType` is null/undefined), which callers
* should treat as "every knob applies".
*/
export function getProviderCapabilities(
providerType: string | null | undefined,
): ProviderCapabilities | null {
if (!providerType) return null;
return PROVIDER_CAPABILITIES[providerType] ?? DEFAULT_EXTERNAL_CAPABILITIES;
}