Qwen2.5-VL 72B Instruct vs Step 2 (16K) vs Command R+
Qwen2.5-VL 72B Instruct comes out ahead, 30 to 22 and 11 on our weighted score, and it is the cheaper option too.
- Our pick
Alibaba (Qwen)Qwen2.5-VL 72B Instruct
30/100- ECI—
- Price$2.80 / $8.40
- Context131K
StepFunStep 2 (16K)
11/100- ECI—
- Price$5.21 / $16.44
- Context16K
CohereCommand R+
22/100- ECI—
- Price$2.50 / $10.00
- Context128K
Qwen2.5-VL 72B Instruct is our pick
Qwen2.5-VL 72B Instruct is the better all-round choice, scoring 30/100 against Command R+ (22) and Step 2 (16K) (11). It leads on inputs & features. 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.
- CapabilityNot enough dataNo independent benchmark covers every model here yet
- Lowest priceQwen2.5-VL 72B InstructQwen2.5-VL 72B Instruct $4.20 · Command R+ $4.38 · Step 2 (16K) $8.02 per 1M tokens (3:1 blend)
- Longest contextQwen2.5-VL 72B InstructQwen2.5-VL 72B Instruct 131,072 · Command R+ 128,000 · Step 2 (16K) 16,384 tokens
- Widest inputsQwen2.5-VL 72B InstructQwen2.5-VL 72B Instruct: Text, Images · Step 2 (16K): Text · Command R+: Text
- Self-hostingQwen2.5-VL 72B Instruct and Command R+Publishes downloadable weights
| Measure | Weight | Qwen2.5-VL 72B Instruct | Step 2 (16K) | Command R+ |
|---|---|---|---|---|
| Price | 50% | 20 | 7 | 19 |
| Inputs & features | 30% | 50 | 25 | 25 |
| Context window | 20% | 24 | 0 | 24 |
| Overall | 100% | 30/100 | 11/100 | 22/100 |
Left out because at least one model lacks the data: capability. The remaining weights were rescaled.
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | Qwen2.5-VL 72B InstructAlibaba (Qwen) | Step 2 (16K)StepFun | Command R+Cohere |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | — | — | — |
| ECI rank | — | — | — |
| Price per million tokens | |||
| Input | $2.80 | $5.21 | $2.50 (best) |
| Output | $8.40 (best) | $16.44 | $10.00 |
| Cached input | — | $1.04 | — |
| Blended (3:1) | $4.20 (best) | $8.02 | $4.38 |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Official Alibaba API | Official StepFun (Global) API | Official Cohere API |
| Limits | |||
| Context window | 131,072 tokens (best) | 16,384 tokens | 128,000 tokens |
| Max output | 8,192 tokens (best) | 8,192 tokens (best) | 4,000 tokens |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | Yes | No | No |
| PDFs | No | No | No |
| Audio | No | No | No |
| Video | No | No | No |
| Reasoning | No | No | No |
| Tool calling | Yes | Yes | Yes |
| Structured output | No | No | No |
| Availability | |||
| Weights | Open | Proprietary | Open |
| API model ID | qwen2-5-vl-72b-instruct | step-2-16k | command-r-plus-08-2024 |
| API providers | 1 | 1 | 7 (best) |
| Released | Sep 2024 | Jan 1, 2025 | Aug 30, 2024 |
| Knowledge cutoff | Apr 2024 | Jun 2024 | Jun 1, 2024 |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
- Qwen2.5-VL 72B Instruct$44.80
- Step 2 (16K)$84.98
- Command R+$45.00
Which should you choose?
Which is better: Qwen2.5-VL 72B Instruct, Step 2 (16K) or Command R+?
Qwen2.5-VL 72B Instruct is the better all-round choice, scoring 30/100 against Command R+ (22) and Step 2 (16K) (11). It leads on inputs & features. 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.
Which is cheaper, Qwen2.5-VL 72B Instruct, Step 2 (16K) or Command R+?
Qwen2.5-VL 72B Instruct is cheaper at $2.80 input / $8.40 output per million tokens (official Alibaba API price). Command R+ costs $2.50 input / $10.00 output per million tokens (official Cohere API price); Step 2 (16K) costs $5.21 input / $16.44 output per million tokens (official StepFun (Global) API price). At a typical mix of three input tokens to one output token, that is $4.20 per million tokens for Qwen2.5-VL 72B Instruct versus $4.38 for Command R+ (1× as much) and $8.02 for Step 2 (16K) (1.9× as much).
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. Qwen2.5-VL 72B Instruct has not been scored yet, Step 2 (16K) has not been scored yet and Command R+ has not been scored yet.
Which is better for coding?
There are no published SWE-bench Verified results for Qwen2.5-VL 72B Instruct, Step 2 (16K) and Command R+ 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.
Which has the bigger context window?
Qwen2.5-VL 72B Instruct has the largest context window at 131,072 tokens, against 128,000 for Command R+ and 16,384 for Step 2 (16K). Maximum output per response: Qwen2.5-VL 72B Instruct up to 8,192, Step 2 (16K) up to 8,192, Command R+ up to 4,000 tokens.
Which can read images, PDFs, audio or video?
Qwen2.5-VL 72B Instruct accepts text and images; Step 2 (16K) accepts text; Command R+ accepts text. Qwen2.5-VL 72B Instruct handles the widest range of inputs.
Are any of these open source?
Qwen2.5-VL 72B Instruct and Command R+ publishes its weights and can be self-hosted; Step 2 (16K) is proprietary.
Which is newer?
Step 2 (16K) is the newest, released Jan 1, 2025. Qwen2.5-VL 72B Instruct came out Sep 2024; Command R+ came out Aug 30, 2024. Knowledge cutoff: Qwen2.5-VL 72B Instruct Apr 2024, Step 2 (16K) Jun 2024, Command R+ Jun 1, 2024.
How do you decide the winner?
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.
