GLOBALATLAS AI · DEDICATED INFERENCE

Mistral-7B-Instruct-v0.3

Mistral-7B-Instruct-v0.3: habayn inference oo iskiis loo marti-geliyo, weights BF16, context 8192-token ah, xusuus GPU iyo xulashada server-ka.

Qaabeynta

Model qoraal oo instructions-ka lagu habeeyey oo leh tokenizer-ka v3. Function calling wuxuu u baahan yahay chat template iswaafaqsan iyo fulinta qalabka dhinaca application-ka.

Moodel: mistralai/Mistral-7B-Instruct-v0.3
Parameters / weights: 7.25B · — GB
Qaabka weights-ka: BF16
Context / concurrerncy: 8192 / 1
Xadka context-ka checkpoint: 32768
Xusuusta GPU ee la qiyaasay: Requires an updated assessment
RAM / disk la qorsheeyey: ≥ 32 GB / ≥ 100 GB
Runtime serving: vLLM · HTTP API
Shati: Apache-2.0

Qiyaastu waxay ku dartaa weights BF16, KV cache, goob shaqo 2 GB ah iyo kayd xad-dhaaf ah oo 15% GPU ah. RAM iyo disk-ga waa shuruudo qorsheyn GlobalAtlas, ma aha minima rasmiga ah ee horumariyaha. Nooca disk-ga, drivers-ka iyo runtime-ka waa la xaqiijiyaa ka hor rakibidda. Miisaaniyadda serving-ka ee 8192-token waxay ku jirtaa gelinta iyo soo saarista; context dheer ama codsiyo isku mar badan waxay u baahan yihiin cabbir beddelid iyo load test.

Kaarka asalka ah ee model-ka

Dooro kaabayaashaaga

Dooro hardware-ka, ka dibna ku xaqiiji helitaanka iyo rakibidda model-ka khabiir. Dalabka server-ka oo keliya ma rakibo model-ka.

Waxaan kula dooran doonnaa habayn iswaafaqsan model-kan.

Codso dejinta oo dhamaystiran

Ku dar model-ka, gobolka iyo culeyska la filayo. Khubaradayadu waxay xaqiijin doonaan hardware-ka, rakibidda, ilaalinta gelitaanka iyo qiimaha kama dambaysta ah.

Kala hadal Atlas →

Codso caawimo khabiir →

U dir shuruudahaaga khubarada GlobalAtlas si loo sameeyo dib-u-eegis habayn gaar ah. Waxay hubin doonaan hardware-ka, iswaafajinta iyo shuruudaha, ka dibna waxay kaga jawaabi doonaan iimayl.