Baichuan M2 32B vs Command R7B Arabic vs Phi-4-multimodal-instruct
Phi-4-multimodal-instruct comes out ahead, 72 to 65 and 58 on our weighted score, though Command R7B Arabic is 18% cheaper per token.
BaichuanBaichuan M2 32B
58/100- ECI—
- Price$0.07 / $0.07
- Context131K
CohereCommand R7B Arabic
65/100- ECI—
- Price$0.037 / $0.15
- Context128K
- Our pick
MicrosoftPhi-4-multimodal-instruct
72/100- ECI—
- Price$0.07 / $0.11
- Context131K
Phi-4-multimodal-instruct is our pick
Phi-4-multimodal-instruct is the better all-round choice, scoring 72/100 against Command R7B Arabic (65) and Baichuan M2 32B (58). 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 priceCommand R7B ArabicCommand R7B Arabic $0.066 · Baichuan M2 32B $0.07 · Phi-4-multimodal-instruct $0.08 per 1M tokens (3:1 blend)
- Longest contextBaichuan M2 32B and Phi-4-multimodal-instructBaichuan M2 32B 131,072 · Phi-4-multimodal-instruct 131,072 · Command R7B Arabic 128,000 tokens
- Widest inputsPhi-4-multimodal-instructBaichuan M2 32B: Text · Command R7B Arabic: Text · Phi-4-multimodal-instruct: Text, Images, Audio
- Self-hostingAll open weightsEvery model here can be downloaded and run on your own hardware
| Measure | Weight | Baichuan M2 32B | Command R7B Arabic | Phi-4-multimodal-instruct |
|---|---|---|---|---|
| Price | 50% | 100 | 100 | 98 |
| Inputs & features | 30% | 10 | 35 | 60 |
| Context window | 20% | 25 | 24 | 25 |
| Overall | 100% | 58/100 | 65/100 | 72/100 |
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 | Baichuan M2 32BBaichuan | Command R7B ArabicCohere | Phi-4-multimodal-instructMicrosoft |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | — | — | — |
| ECI rank | — | — | — |
| Price per million tokens | |||
| Input | $0.07 | $0.037 (best) | $0.07 |
| Output | $0.07 (best) | $0.15 | $0.11 |
| Cached input | — | — | — |
| Blended (3:1) | $0.07 | $0.066 (best) | $0.08 |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Median of 1 providers | Official Cohere API | Median of 1 providers |
| Limits | |||
| Context window | 131,072 tokens (best) | 128,000 tokens | 131,072 tokens (best) |
| Max output | 131,072 tokens (best) | 4,000 tokens | 131,072 tokens (best) |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | No | No | Yes |
| PDFs | No | No | No |
| Audio | No | No | Yes |
| Video | No | No | No |
| Reasoning | Yes | No | No |
| Tool calling | No | Yes | Yes |
| Structured output | No | Yes | No |
| Availability | |||
| Weights | OpenApache-2.0 | Open | OpenMIT |
| API model ID | — | command-r7b-arabic-02-2025 | — |
| API providers | 1 | 1 | 1 |
| Released | Aug 13, 2025 | Feb 27, 2025 | Feb 26, 2025 |
| Knowledge cutoff | — | Jun 1, 2024 | Jun 2024 |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
- Baichuan M2 32B$0.84
- Command R7B Arabic$0.675
- Phi-4-multimodal-instruct$0.92
Which should you choose?
Which is better: Baichuan M2 32B, Command R7B Arabic or Phi-4-multimodal-instruct?
Phi-4-multimodal-instruct is the better all-round choice, scoring 72/100 against Command R7B Arabic (65) and Baichuan M2 32B (58). 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, Baichuan M2 32B, Command R7B Arabic or Phi-4-multimodal-instruct?
Command R7B Arabic is cheaper at $0.037 input / $0.15 output per million tokens (official Cohere API price). Baichuan M2 32B costs $0.07 input / $0.07 output per million tokens (median across 1 API provider); Phi-4-multimodal-instruct costs $0.07 input / $0.11 output per million tokens (median across 1 API provider). At a typical mix of three input tokens to one output token, that is $0.066 per million tokens for Command R7B Arabic versus $0.07 for Baichuan M2 32B (1.1× as much) and $0.08 for Phi-4-multimodal-instruct (1.2× as much).
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. Baichuan M2 32B has not been scored yet, Command R7B Arabic has not been scored yet and Phi-4-multimodal-instruct has not been scored yet.
Which is better for coding?
There are no published SWE-bench Verified results for Baichuan M2 32B, Command R7B Arabic and Phi-4-multimodal-instruct yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. Note that Baichuan M2 32B does not support tool calling, which most coding agents need.
Which has the bigger context window?
Baichuan M2 32B and Phi-4-multimodal-instruct have the largest context windows (131,072 and 131,072 tokens), against 128,000 for Command R7B Arabic. Maximum output per response: Baichuan M2 32B up to 131,072, Command R7B Arabic up to 4,000, Phi-4-multimodal-instruct up to 131,072 tokens.
Which can read images, PDFs, audio or video?
Baichuan M2 32B accepts text; Command R7B Arabic accepts text; Phi-4-multimodal-instruct accepts text, images and audio. Phi-4-multimodal-instruct handles the widest range of inputs.
Are any of these open source?
Yes, all three publish their weights (Apache-2.0 and MIT), so you can self-host them.
Which is newer?
Baichuan M2 32B is the newest, released Aug 13, 2025. Command R7B Arabic came out Feb 27, 2025; Phi-4-multimodal-instruct came out Feb 26, 2025. Knowledge cutoff: Command R7B Arabic Jun 1, 2024, Phi-4-multimodal-instruct Jun 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 Nov 2, 2026.
