Command A vs Qwen3-Coder 480B-A35B Instruct vs Step 1 (32K)

Qwen3-Coder 480B-A35B Instruct is the better all-round choice, scoring 28/100 against Command A (24) and Step 1 (32K) (18). It leads on price. The score weighs price 50%, inputs & features 30%, context window 20%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.

Qwen3-Coder 480B-A35B Instruct is cheaper at $1.50 input / $7.50 output per million tokens (official Alibaba API price). Step 1 (32K) costs $2.05 input / $9.59 output per million tokens (official StepFun (Global) API price); Command A costs $2.50 input / $10.00 output per million tokens (official Cohere API price). At a typical mix of three input tokens to one output token, that is $3.00 per million tokens for Qwen3-Coder 480B-A35B Instruct versus $3.94 for Step 1 (32K) (1.3× as much) and $4.38 for Command A (1.5× as much).

There is no independent benchmark that covers all three models yet. Command A has not been scored yet, Qwen3-Coder 480B-A35B Instruct has not been scored yet and Step 1 (32K) has not been scored yet.

There are no published SWE-bench Verified results for Command A, Qwen3-Coder 480B-A35B Instruct and Step 1 (32K) yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. All three support tool calling for agent workflows.

Qwen3-Coder 480B-A35B Instruct has the largest context window at 262,144 tokens, against 256,000 for Command A and 32,768 for Step 1 (32K). Maximum output per response: Command A up to 8,000, Qwen3-Coder 480B-A35B Instruct up to 65,536, Step 1 (32K) up to 32,768 tokens.

Command A accepts text; Qwen3-Coder 480B-A35B Instruct accepts text; Step 1 (32K) accepts text. They handle the same number of input types.

Command A and Qwen3-Coder 480B-A35B Instruct publishes its weights and can be self-hosted; Step 1 (32K) is proprietary.

Qwen3-Coder 480B-A35B Instruct is the newest, released Apr 2025. Command A came out Mar 13, 2025; Step 1 (32K) came out Jan 1, 2025. Knowledge cutoff: Command A Jun 1, 2024, Qwen3-Coder 480B-A35B Instruct Apr 2025, Step 1 (32K) Jun 2024.

Each model gets a 0–100 score on capability (50%, independent benchmark results); price (25%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (15%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (10%, maximum tokens per request, log scale). Dimensions missing for any model are dropped and the remaining weights rescaled, so every model is judged on the same evidence. Specs and prices come from public model listings and the labs’ own API pages; capability scores come from independent benchmark runs. Data updated Oct 4, 2026.