MiniMax-M3 vs Qwen3.5 4B vs Qwen3.7 Plus
Qwen3.7 Plus comes out ahead, 87 to 70 and 57 on our weighted score, though Qwen3.5 4B is 15× cheaper per token.
Price in rating: Excluded · Include price
MiniMax
MiniMax-M3
70/100- ECI147.9
- Price$0.30 / $1.20
- Context1.05M
Alibaba (Qwen)
Qwen3.5 4B
57/100- ECI—
- Price$0.04 / $0.07
- Context262K
- Our pick
Alibaba (Qwen)
Qwen3.7 Plus
87/100- ECI148.0
- Price$0.40 / $1.60
- Context1M
Qwen3.7 Plus is our pick
Qwen3.7 Plus is the better all-round choice, scoring 87/100 against MiniMax-M3 (70) and Qwen3.5 4B (57). It leads on capability and inputs & features. The score weighs capability 67%, inputs & features 20%, context window 13%. Capability uses the average of shared benchmarks (OTIS Mock AIME 2024–2025), because Qwen3.5 4B has no Capabilities Index score yet.
- CapabilityQwen3.7 PlusShared benchmarks: Qwen3.7 Plus 93.3% · MiniMax-M3 71.1% · Qwen3.5 4B 55.8%
- Lowest priceQwen3.5 4BQwen3.5 4B $0.048 · MiniMax-M3 $0.525 · Qwen3.7 Plus $0.70 per 1M tokens (3:1 blend)
- Longest contextMiniMax-M3MiniMax-M3 1,048,576 · Qwen3.7 Plus 1,000,000 · Qwen3.5 4B 262,144 tokens
- Widest inputsSame inputsMiniMax-M3: Text, Images, Video · Qwen3.5 4B: Text, Images, Video · Qwen3.7 Plus: Text, Images, Video
- Self-hostingMiniMax-M3 and Qwen3.5 4BPublishes downloadable weights (Apache-2.0)
| Measure | Weight | MiniMax-M3 | Qwen3.5 4B | Qwen3.7 Plus |
|---|---|---|---|---|
| CapabilityShared benchmarks | 67% | 71 | 56 | 93 |
| Inputs & features | 20% | 70 | 70 | 80 |
| Context window | 13% | 65 | 43 | 64 |
| Overall | 100% | 70/100 | 57/100 | 87/100 |
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | |||
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | 147.9 | — | 148.0 (best) |
| ECI rank | #77 of 158 | — | #75 of 158 (best) |
| GPQA DiamondGraduate-level science questions | 90.9% (best) | — | 87.9% |
| FrontierMath Tiers 1–3Research-level mathematics | — | — | 34.4% |
| OTIS Mock AIME 2024–2025Competition mathematics | 71.1% | 55.8% | 93.3% (best) |
| Price per million tokens | |||
| Input | $0.30 | $0.04 (best) | $0.40 |
| Output | $1.20 | $0.07 (best) | $1.60 |
| Cached input | $0.06 (best) | — | $0.08 |
| Blended (3:1) | $0.525 | $0.048 (best) | $0.70 |
| Long-context rate | Over 512K: $0.60 / $2.40 | Same rate | Over 256K: $1.20 / $4.80 |
| Price source | Official MiniMax (minimax.io) API | Median of 3 providers | Official Alibaba API |
| Limits | |||
| Context window | 1,048,576 tokens (best) | 262,144 tokens | 1,000,000 tokens |
| Max output | 512,000 tokens (best) | 81,920 tokens | 64,000 tokens |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | Yes | Yes | Yes |
| PDFs | No | No | No |
| Audio | No | No | No |
| Video | Yes | Yes | Yes |
| Reasoning | Yes | Yes | Yes |
| Tool calling | Yes | Yes | Yes |
| Structured output | No | No | Yes |
| Availability | |||
| Weights | Open | OpenApache-2.0 | Proprietary |
| API model ID | MiniMax-M3 | — | qwen3.7-plus |
| API providers | 52 (best) | 5 | 28 |
| Released | Jun 1, 2026 | Mar 2, 2026 | Jun 2, 2026 |
| Knowledge cutoff | — | — | Apr 2025 |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
MiniMax-M3$5.40
Qwen3.5 4B$0.54
Qwen3.7 Plus$7.20
Which should you choose?
Which is better: MiniMax-M3, Qwen3.5 4B or Qwen3.7 Plus?
Qwen3.7 Plus is the better all-round choice, scoring 87/100 against MiniMax-M3 (70) and Qwen3.5 4B (57). It leads on capability and inputs & features. The score weighs capability 67%, inputs & features 20%, context window 13%. Capability uses the average of shared benchmarks (OTIS Mock AIME 2024–2025), because Qwen3.5 4B has no Capabilities Index score yet.
Which is cheaper, MiniMax-M3, Qwen3.5 4B or Qwen3.7 Plus?
Qwen3.5 4B is cheaper at $0.04 input / $0.07 output per million tokens (median across 3 API providers). MiniMax-M3 costs $0.30 input / $1.20 output per million tokens (official MiniMax (minimax.io) API price); Qwen3.7 Plus costs $0.40 input / $1.60 output per million tokens (official Alibaba API price). At a typical mix of three input tokens to one output token, that is $0.048 per million tokens for Qwen3.5 4B versus $0.525 for MiniMax-M3 (11× as much) and $0.70 for Qwen3.7 Plus (15× as much).
Which scores higher on benchmarks?
Not every model here has a Capabilities Index score, so we compare the average of shared benchmarks (OTIS Mock AIME 2024–2025): Qwen3.7 Plus 93.3%, MiniMax-M3 71.1% and Qwen3.5 4B 55.8%. On individual benchmarks: OTIS Mock AIME 2024–2025 — Qwen3.7 Plus 93.3%, MiniMax-M3 71.1%, Qwen3.5 4B 55.8%.
Which is better for coding?
There are no published SWE-bench Verified results for MiniMax-M3, Qwen3.5 4B and Qwen3.7 Plus yet, so there is no like-for-like coding score. On overall capability, Qwen3.7 Plus leads, which tends to carry over to coding, but test on your own codebase. All three support tool calling for agent workflows.
Which has the bigger context window?
MiniMax-M3 has the largest context window at 1,048,576 tokens, against 1,000,000 for Qwen3.7 Plus and 262,144 for Qwen3.5 4B. Maximum output per response: MiniMax-M3 up to 512,000, Qwen3.5 4B up to 81,920, Qwen3.7 Plus up to 64,000 tokens.
Which can read images, PDFs, audio or video?
MiniMax-M3 accepts text, images and video; Qwen3.5 4B accepts text, images and video; Qwen3.7 Plus accepts text, images and video. They handle the same number of input types.
Are any of these open source?
MiniMax-M3 and Qwen3.5 4B publishes its weights (Apache-2.0) and can be self-hosted; Qwen3.7 Plus is proprietary.
Which is newer?
Qwen3.7 Plus is the newest, released Jun 2, 2026. MiniMax-M3 came out Jun 1, 2026; Qwen3.5 4B came out Mar 2, 2026. Knowledge cutoff: Qwen3.7 Plus Apr 2025.
How do you decide the winner?
This rating uses capability (67%, independent benchmark results); inputs & features (20%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (13%, 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 18, 2026.