qwen3.5-122b-a10b API — compare providers

The same model (qwen3.5-122b-a10b) is served by 5 providers; SiliconFlow is the cheapest at $2.34/1M blended — the priciest (Novita AI) costs 54% more. On quality, qwen3.5-122b-a10b scores 1417 on the LMArena (Chatbot Arena) — so the cheapest endpoint is also the best value per dollar here.

LicenseApache 2.0open weights · permissive · terms
Knowledge cutoffSeptember 2025source · ranked
Quality (Elo)1417LMArena (Chatbot Arena) · source
Intelligence32.9Artificial Analysis Intelligence Index · source · ranked

Quality and intelligence scores are attached to the model itself — identical whichever provider serves it. Verified 2026-09-04.

RankProviderInput /1MOutput /1MBlended /1MValue (Elo/$)
1SiliconFlow$0.26$2.08$2.34 ← cheapest606 ← best value
2Alibaba$0.26$2.08$2.34606
3DeepInfra$0.29$2.4$2.69527
4AtlasCloud$0.3$2.4$2.7525
5Novita AI$0.4$3.2$3.6394

Value = model Elo ÷ blended price per 1M tokens. Since the model's quality is identical across providers, the cheapest endpoint is also the best value per dollar.

Prices are per 1M tokens (USD). "Blended" = input + output, for coarse ranking. Same underlying model, 5 serving providers, 54% spread top to bottom. Context window: 262,144 tokens (235,929 max output) — see how it ranks in biggest context windows. See how this compares across the catalogue in same model, different price, or find the best quality per dollar in the value ranking.

Cost per 1,000 requests by workload

What each provider actually bills for a representative job, not just the sticker price.

ProviderChatbot
1000 in / 500 out
RAG / long context
8000 in / 500 out
Batch summarize
4000 in / 1000 out
SiliconFlow$1.30$3.12$3.12
Alibaba$1.30$3.12$3.12
DeepInfra$1.49$3.52$3.56
AtlasCloud$1.50$3.60$3.60
Novita AI$2.00$4.80$4.80

Estimate your own workload

Input tokens/request:   Output tokens/request:   Requests:

ProviderEstimated cost (USD)

Self-host / quantized builds

Open weights under Apache 2.0 — you can run qwen3.5-122b-a10b on your own hardware instead of paying per token. Even quantized to int4 it needs roughly 73.2 GB of VRAM — a multi-GPU workstation or a large-memory Mac, not a typical desktop. Formats: GGUF for llama.cpp / Ollama / LM Studio, AWQ and GPTQ for GPU serving, MLX for Apple silicon.

Links search the live model hubs (Hugging Face, Ollama) so they track new builds as they appear — we don't host weights. Which formats exist depends on what the model owner and community have published.

→ See VRAM, GPU and cost vs API for qwen3.5-122b-a10b, or the full self-host break-even calculator.