Qwen3 235B-A22B Instruct 2507

qwen/qwen3-235b-a22b-instruct-2507

Qwen (Alibaba)· 2025-07-21· 262K ctx· apache-2.0

open weightstool callstructured quant: unknown

spread (output, p90 ÷ p10)

4.4×

$0.550 – $3.08 · 10 providers · max÷min 5.6×

10 providers priced · 10 offerings · 1 omitted (stale)

Including routers & resellers

Cheapest / median / spread use providers that serve the model on their own capacity: first-party lab, hyperscaler, inference host.

Provider classes in the stats:

Blended $/M assumptionsInput : output: 1Cache-hit share of input%

blended = 90.9% input × [(1 − 70% cache) × input + 70% cache × cache-read] + 9.1% × output + cache writes (30% of uncached input) where a write price is published

A provider that publishes no cache price is blended with cache share 0 (all input at the input price) and labelled “no cache price published” — never penalised against identical in/out prices.

Provider	Input	Output	Cache rd	Blended
submodel unclassified	$0.200	$0.300	—	$0.209no cache price published
OpenRouter router / aggregator	$0.0875	$0.350	$0.0175	$0.0668
Cortecs router / aggregator	$0.0690	$0.455	$0.0180	$0.0716
NanoGPT router / aggregator	$0.130	$0.500	$0.0650	$0.122
Deep Infra inference host	$0.0900	$0.550	—	$0.132no cache price published
LLM Gateway router / aggregator	$0.0900	$0.580	—	$0.135no cache price published
NovitaAI inference host	$0.0900	$0.580	—	$0.135no cache price published
Meganova unclassified	$0.0900	$0.600	—	$0.136no cache price published
Merge Gateway router / aggregator	$0.100	$0.600	$0.0500	$0.114
Abacus router / aggregator	$0.130	$0.600	—	$0.173no cache price published
Nebius Token Factory inference host	$0.200	$0.600	—	$0.236no cache price published
Together AI inference host	$0.200	$0.600	—	$0.236no cache price published
Venice AI inference host	$0.150	$0.750	—	$0.205no cache price published
Jiekou.AI reseller	$0.150	$0.800	—	$0.209no cache price published
Crusoe inference host	$0.220	$0.800	$0.110	$0.203
Vercel AI Gateway router / aggregator	$0.220	$0.880	—	$0.280no cache price published
Amazon Bedrock hyperscaler	$0.220	$0.880	—	$0.280no cache price published
Eden AI router / aggregator	$0.230	$0.920	—	$0.293no cache price published
302.AI reseller	$0.29	$1.14	—	$0.37no cache price published
Pioneer unclassified	$1.20	$1.20	$1.20	$1.30
Scaleway inference host	$0.75	$2.25	—	$0.89no cache price published
Hugging Face inference host	$0.85	$2.56	—	$1.01no cache price published
GreenPT inference host	$1.03	$3.08	—	$1.21no cache price published

Stale offerings (not in the spread) · 1

Last priced more than 3 days before the latest snapshot — shown for reference, never mixed into cheapest, median or spread.

NanoGPT qwen/Qwen3-235B-A22B-Instruct-2507pay per token

Subscription plans (flat fee, not per-token) · 1

Flat-fee plan — the per-token price is bundled into a subscription, so $0/token is not free. Excluded from cheapest, median and spread.

DevPass (LLM Gateway) qwen3-235b-a22b-instruct-2507subscription plan

Free tiers (rate-limited) · 1

No payment required; rate limits and quotas apply (not tracked yet). Excluded from cheapest, median and spread.

ModelScope Qwen/Qwen3-235B-A22B-Instruct-2507China-hosted endpoint (modelscope.cn)ModelScope account + access tokenfree tier

Connect

Provider (default: cheapest output among direct providers)

$0.550 /M · inference host

opencode.json

{
  "provider": {
    "deepinfra": {
      "npm": "@ai-sdk/deepinfra",
      "models": {
        "Qwen/Qwen3-235B-A22B-Instruct-2507": {}
      }
    }
  }
}

Set DEEPINFRA_API_KEY in your environment. Generated — verify against the OpenCode docs.

Output $/M over time7-day preview

Median line, min–max band. Pre-2026-09 points are catalog-recorded dates from models.dev git history, not provider effective dates.

Data: models.dev (MIT), snapshot 2026-09-20. Prices are vendor-published; verify before purchase. Methodology.