@@ -2357,34 +2357,6 @@ describe.skip("InferenceClient", () => {
23572357 status : "live" ,
23582358 task : "conversational" ,
23592359 } ,
2360- "Qwen/Qwen3-Coder-480B-A35B-Instruct" : {
2361- provider : "baseten" ,
2362- hfModelId : "Qwen/Qwen3-Coder-480B-A35B-Instruct" ,
2363- providerId : "Qwen/Qwen3-Coder-480B-A35B-Instruct" ,
2364- status : "live" ,
2365- task : "conversational" ,
2366- } ,
2367- "moonshotai/Kimi-K2-Instruct-0905" : {
2368- provider : "baseten" ,
2369- hfModelId : "moonshotai/Kimi-K2-Instruct-0905" ,
2370- providerId : "moonshotai/Kimi-K2-Instruct-0905" ,
2371- status : "live" ,
2372- task : "conversational" ,
2373- } ,
2374- "deepseek-ai/DeepSeek-V3.1" : {
2375- provider : "baseten" ,
2376- hfModelId : "deepseek-ai/DeepSeek-V3.1" ,
2377- providerId : "deepseek-ai/DeepSeek-V3.1" ,
2378- status : "live" ,
2379- task : "conversational" ,
2380- } ,
2381- "deepseek-ai/DeepSeek-V3-0324" : {
2382- provider : "baseten" ,
2383- hfModelId : "deepseek-ai/DeepSeek-V3-0324" ,
2384- providerId : "deepseek-ai/DeepSeek-V3-0324" ,
2385- status : "live" ,
2386- task : "conversational" ,
2387- } ,
23882360 } ;
23892361
23902362 it ( "chatCompletion - Qwen3 235B Instruct" , async ( ) => {
@@ -2402,63 +2374,6 @@ describe.skip("InferenceClient", () => {
24022374 }
24032375 } ) ;
24042376
2405- it ( "chatCompletion - Qwen3 Coder 480B" , async ( ) => {
2406- const res = await client . chatCompletion ( {
2407- model : "Qwen/Qwen3-Coder-480B-A35B-Instruct" ,
2408- provider : "baseten" ,
2409- messages : [ { role : "user" , content : "Write a simple Python function to add two numbers" } ] ,
2410- max_tokens : 100 ,
2411- } ) ;
2412- if ( res . choices && res . choices . length > 0 ) {
2413- const completion = res . choices [ 0 ] . message ?. content ;
2414- expect ( completion ) . toBeDefined ( ) ;
2415- expect ( typeof completion ) . toBe ( "string" ) ;
2416- expect ( completion ) . toMatch ( / d e f .* a d d / i) ;
2417- }
2418- } ) ;
2419-
2420- it ( "chatCompletion - Kimi K2 Instruct" , async ( ) => {
2421- const res = await client . chatCompletion ( {
2422- model : "moonshotai/Kimi-K2-Instruct-0905" ,
2423- provider : "baseten" ,
2424- messages : [ { role : "user" , content : "What is the capital of Japan?" } ] ,
2425- temperature : 0.1 ,
2426- } ) ;
2427- if ( res . choices && res . choices . length > 0 ) {
2428- const completion = res . choices [ 0 ] . message ?. content ;
2429- expect ( completion ) . toBeDefined ( ) ;
2430- expect ( typeof completion ) . toBe ( "string" ) ;
2431- expect ( completion ) . toMatch ( / t o k y o / i) ;
2432- }
2433- } ) ;
2434-
2435- it ( "chatCompletion - DeepSeek V3.1" , async ( ) => {
2436- const res = await client . chatCompletion ( {
2437- model : "deepseek-ai/DeepSeek-V3.1" ,
2438- provider : "baseten" ,
2439- messages : [ { role : "user" , content : "Complete this sentence with words, one plus one is equal " } ] ,
2440- } ) ;
2441- if ( res . choices && res . choices . length > 0 ) {
2442- const completion = res . choices [ 0 ] . message ?. content ;
2443- expect ( completion ) . toContain ( "two" ) ;
2444- }
2445- } ) ;
2446-
2447- it ( "chatCompletion - DeepSeek V3" , async ( ) => {
2448- const res = await client . chatCompletion ( {
2449- model : "deepseek-ai/DeepSeek-V3-0324" ,
2450- provider : "baseten" ,
2451- messages : [ { role : "user" , content : "What is 2 * 3?" } ] ,
2452- temperature : 0.1 ,
2453- } ) ;
2454- if ( res . choices && res . choices . length > 0 ) {
2455- const completion = res . choices [ 0 ] . message ?. content ;
2456- expect ( completion ) . toBeDefined ( ) ;
2457- expect ( typeof completion ) . toBe ( "string" ) ;
2458- expect ( completion ) . toMatch ( / ( s i x | 6 ) / i) ;
2459- }
2460- } ) ;
2461-
24622377 it ( "chatCompletion stream - Qwen3 235B" , async ( ) => {
24632378 const stream = client . chatCompletionStream ( {
24642379 model : "Qwen/Qwen3-235B-A22B-Instruct-2507" ,
@@ -2483,30 +2398,6 @@ describe.skip("InferenceClient", () => {
24832398 expect ( fullResponse . length ) . toBeGreaterThan ( 0 ) ;
24842399 expect ( fullResponse ) . toMatch ( / 1 .* 2 .* 3 / ) ;
24852400 } ) ;
2486-
2487- it ( "chatCompletion stream - DeepSeek V3.1" , async ( ) => {
2488- const stream = client . chatCompletionStream ( {
2489- model : "deepseek-ai/DeepSeek-V3.1" ,
2490- provider : "baseten" ,
2491- messages : [ { role : "user" , content : "Complete the equation 1 + 1 = , just the answer" } ] ,
2492- stream : true ,
2493- } ) as AsyncGenerator < ChatCompletionStreamOutput > ;
2494-
2495- let fullResponse = "" ;
2496- for await ( const chunk of stream ) {
2497- if ( chunk . choices && chunk . choices . length > 0 ) {
2498- const content = chunk . choices [ 0 ] . delta ?. content ;
2499- if ( content ) {
2500- fullResponse += content ;
2501- }
2502- }
2503- }
2504-
2505- // Verify we got a meaningful response
2506- expect ( fullResponse ) . toBeTruthy ( ) ;
2507- expect ( fullResponse . length ) . toBeGreaterThan ( 0 ) ;
2508- expect ( fullResponse ) . toMatch ( / ( t w o | 2 ) / i) ;
2509- } ) ;
25102401 } ,
25112402 TIMEOUT
25122403 ) ;
0 commit comments