MiniMax M3 vs Qwen3.7 Plus
MiniMax M3 has been retired from our catalogue. Its figures below are the last published rates; calls to it are no longer served, while the other model in this comparison is.
Which one, when
These two are close on paper: both minimax-m3 and qwen3.7-plus offer a 1000000-token context, text/image/video input with text output, and the same chat, code, reasoning, tools and long-context flags, with thinking that can be disabled, and both released 2026-06-01. The separation is rate card and output ceiling: qwen3.7-plus bills about 1.33x minimax-m3 on input ($0.4 vs $0.3), output ($1.6 vs $1.2) and cache reads ($0.08 vs $0.06), while minimax-m3 allows 524288 max output tokens against 65536, or 8x the room. Pick minimax-m3 for cheaper runs and very long single generations; pick qwen3.7-plus if you prefer Alibaba as the vendor.
Benchmarks
MiniMax M3: the vendor has not published benchmark scores.
Vendor-published: Alibaba (Qwen) Alibaba (Qwen) Alibaba (Qwen) Alibaba (Qwen) Alibaba (Qwen) Alibaba (Qwen) Alibaba (Qwen) Anthropic Anthropic Anthropic Anthropic Anthropic ByteDance DeepSeek DeepSeek Google Google Google Google MiniMax Moonshot OpenAI OpenAI OpenAI OpenAI OpenAI Tencent Z.ai Z.ai Z.ai Z.ai
Pricing
| MiniMax M3 | Qwen3.7 Plus | Δ | |
|---|---|---|---|
| Input / 1M tokens | $0.3 | $0.4 | 0.75× |
| Output / 1M tokens | $1.2 | $1.6 | 0.75× |
| Cache read / 1M tokens | $0.06 | $0.08 | 0.75× |
| Cache write | no separate charge | 1.25x | - |
Rates from the live catalogue at build time; each model page carries the current rate card. The main table shows base rates. Prompt length can change the rate for the entire request. See each model’s full rate card.
Qwen3.7 Plus · Rates by prompt length
The whole request uses the rate for its input-length tier. Prices are USD per million tokens.
| Input tokens | Input | Output | Thinking input |
|---|---|---|---|
| 0 - 256,000 | $0.4 | $1.6 | $1.6 |
| 256,001 - 1,000,000 | $1.2 | $4.8 | $4.8 |
Where they sit · input price per 1M tokens across all 91 chat models on this billing unit (log scale)
Capabilities
| MiniMax M3 | Qwen3.7 Plus | |
|---|---|---|
| Tool calling | yes | yes |
| Thinking control | configurable | configurable |
| Structured output | - | yes |
| Prompt caching | implicit (automatic) | implicit + explicit |
| Cache lifetime | not published | explicit: 5m, reset on hit |
| Minimum cached prefix | 512 tokens | 1,024 tokens |
Specs
| MiniMax M3 | Qwen3.7 Plus | |
|---|---|---|
| Input modalities | text image video | text image video |
| Output modalities | text | text |
| Released | 2026-06-01 | 2026-06-01 |
| Context window | 1,000,000 | 1,000,000 |
| Max output | 524,288 | 65,536 |
| Thinking parameter |
|
|
| Accepted values | thinking.type
reasoning_split
| enable_thinking
thinking_budget
preserve_thinking
|
| Default | adaptive: thinking on, with the model deciding when extra reasoning helps | on the Qwen3.7 Plus series is hybrid thinking with thinking enabled by default, and preserve_thinking is off |
Specs are transcribed from each vendor’s documentation; a row a vendor does not publish is left out rather than inferred. Full sources: MiniMax M3 · Qwen3.7 Plus
One prompt, both models, measured through the gateway
MiniMax M3 passed · 3 sentences
Reasoning models don't just answer — they first generate extensive internal "thinking" tokens to work through the problem step-by-step, and you're billed for all of those tokens even though users never see them. A query that previously cost ~200 tokens may now cost 2,000–10,000 once reasoning is enabled, multiplying per-query spend by roughly 5–50x while the visible output stayed the same size — that's the doubling. The lever is tuning: cap reasoning effort on simple queries, cache the reasoning traces, and reserve the full reasoning model for only your hardest 10–20% of requests, which typically restores 60–80% of the savings.
out 764 tok (+629 thinking) latency 3.5 s
Qwen3.7 Plus passed · 3 sentences
Enabling reasoning requires the model to generate extensive internal "thinking" tokens to process complex logic before delivering a final answer. Because our vendor charges us per token consumed, this hidden computational overhead effectively doubles our output volume for the exact same number of user requests. Consequently, our costs have doubled because we are now paying for the model's increased processing time to achieve higher-quality, more accurate results.
out 1428 tok (+1345 thinking) latency 25.2 s
Instruction following (exactly three sentences, so countable), audience shaping (CFO register), and the hidden-thinking billing gap the token meters below expose.
MiniMax M3 passed · 8/8 cases
The bug is that the loop starts at index 1 and treats `cur` as the number of *equal transitions* rather than the run length, so a run of *k* consecutive equal items is reported as `k − 1`, and a single-element list (where the loop body never executes) returns 0 instead of 1. A correct version tracks the actual run length by seeding `cur = 1` and resetting it to 1 (not 0) on a mismatch: ```python def longest_run(xs): if not xs: return 0 best = cur = 1 for i in range(1, len(xs)): if xs[i] == xs[i - 1]: cur += 1 else: cur = 1 if cur > best: best = cur return best ``` Now `[5]` returns `1`, `[1,1,1]` returns `3`, and a run like `[1,2,2,2]` that ends at the last index is correctly captured as `3` because `cur` is updated before `best` is compared on every iteration, including the final one.
out 845 tok (+596 thinking) latency 3.5 s
Qwen3.7 Plus passed · 8/8 cases
## Fixed Code ```python def longest_run(xs): if not xs: return 0 best = 1 cur = 1 for i in range(1, len(xs)): if xs[i] == xs[i - 1]: cur += 1 else: cur = 1 best = max(best, cur) return best ``` ## Bug Explanation The function has two related bugs rooted in how it counts run lengths. First, `cur` is initialized to `0` and only incremented on matches, so it counts the number of *transitions* between equal elements rather than the total elements in the run—a run of N identical items only reaches `cur = N − 1`, undercounting by one. This also causes the single-element case to return `0`, since the loop never executes and `cur` never increments. […]
out 2317 tok (+2005 thinking) latency 40.5 s
Whether the fix is actually correct (runnable), explanation density, and token efficiency on a bounded task.
MiniMax M3 passed · 5/5 fields, guidance null
```json { "company": null, "quarter": "Q3", "revenue_usd_m": 412, "growth_pct": null, "guidance": null } ```
out 1002 tok (+955 thinking) latency 3.6 s
Qwen3.7 Plus passed · 5/5 fields, guidance null
{ "company": "Meridian Robotics", "quarter": "September", "revenue_usd_m": 412, "growth_pct": 11, "guidance": null }
out 1807 tok (+1750 thinking) latency 31.6 s
Schema obedience (no invented fields), hallucination pressure (guidance is explicitly withheld), and structured-output path differences.
MiniMax M3 passed · 121 words, 0 banned, 1 question
Introducing PromptCache, our intelligent caching layer for the API gateway. Tired of paying twice for identical requests? The feature stores prompt-response pairs across every connected provider, returning cached results instead of hitting costly endpoints. Deployment requires only a single configuration flag, and semantic similarity detection happens automatically. Latency drops up to 70% on repeated traffic while costs follow the same downward curve. Developers access detailed cache hit metrics, eviction logs, and provider breakdowns through the dashboard. Whether routing to OpenAI, Anthropic, or local models, every cached token saves time and money. The cache respects TTL policies, supports manual purging, and encrypts stored content. Available today for Pro and Enterprise tiers—enable PromptCache on existing endpoints and watch inference bills shrink immediately.
out 2808 tok (+2656 thinking) latency 6.0 s
Qwen3.7 Plus passed · 120 words, 0 banned, 1 question
Introducing PromptCache, our new API gateway feature designed to store and reuse prompts across multiple artificial intelligence providers. Have you ever noticed how repeatedly sending identical queries drains your budget and increases latency? PromptCache solves this by intelligently storing responses at the gateway level. When your application requests the same prompt from a different provider, the system instantly returns the cached result. This drastically reduces API costs and accelerates response times for your users. You can configure custom expiration times and set specific fallback rules for each vendor. You must stop paying twice for the exact same computation. Please upgrade your entire infrastructure today and experience much faster and cheaper integrations without changing a single line of your application code.
out 4453 tok (+4312 thinking) latency 76.8 s
Constraint obedience (word budget, banned-word list, the single question), style fingerprint, and length control.
Switch between them with one line
Both ids are in every tab below; the highlighted pair of lines is the only edit. Same endpoint, same key, same request shape.
from openai import OpenAI
client = OpenAI(
base_url="https://synthorai.io/v1",
api_key="sk-syn-...",
)
resp = client.chat.completions.create(
model="minimax-m3",
# model="qwen3.7-plus", # uncomment this line, comment the one above
messages=[{"role": "user", "content": "Summarize this diff"}],
reasoning_effort="medium",
)
print(resp.choices[0].message.content)import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://synthorai.io/v1",
apiKey: "sk-syn-...",
});
const resp = await client.chat.completions.create({
model: "minimax-m3",
// model: "qwen3.7-plus", // uncomment this line, comment the one above
messages: [{ role: "user", content: "Summarize this diff" }],
reasoning_effort: "medium",
});
console.log(resp.choices[0].message.content);curl https://synthorai.io/v1/chat/completions \
-H "Authorization: Bearer sk-syn-..." \
-H "Content-Type: application/json" \
-d '{
"model": "minimax-m3",
# "model": "qwen3.7-plus", # uncomment this line, comment the one above
"messages": [{"role": "user", "content": "Hello"}],
"reasoning_effort": "medium"
}'package main
import (
"context"
"fmt"
"github.com/openai/openai-go/v3"
"github.com/openai/openai-go/v3/option"
)
func main() {
client := openai.NewClient(
option.WithBaseURL("https://synthorai.io/v1"),
option.WithAPIKey("sk-syn-..."),
)
resp, _ := client.Chat.Completions.New(context.TODO(), openai.ChatCompletionNewParams{
Model: "minimax-m3",
// Model: "qwen3.7-plus", // uncomment this line, comment the one above
Messages: []openai.ChatCompletionMessageParamUnion{
openai.UserMessage("Summarize this diff"),
},
ReasoningEffort: openai.ReasoningEffortMedium,
})
fmt.Println(resp.Choices[0].Message.Content)
}import com.openai.client.OpenAIClient;
import com.openai.client.okhttp.OpenAIOkHttpClient;
import com.openai.models.chat.completions.*;
import com.openai.models.ReasoningEffort;
OpenAIClient client = OpenAIOkHttpClient.builder()
.baseUrl("https://synthorai.io/v1")
.apiKey("sk-syn-...")
.build();
ChatCompletion resp = client.chat().completions().create(
ChatCompletionCreateParams.builder()
.model("minimax-m3")
// .model("qwen3.7-plus") // uncomment this line, comment the one above
.addUserMessage("Summarize this diff")
.reasoningEffort(ReasoningEffort.MEDIUM)
.build());
System.out.println(resp.choices().get(0).message().content().orElse(""));FAQ
Which is cheaper, MiniMax M3 or Qwen3.7 Plus?
The main table shows base rates. Prompt length can change the rate for the entire request. See each model’s full rate card.
Can I A/B test MiniMax M3 against Qwen3.7 Plus without two integrations?
Yes. Both are served through the same OpenAI-compatible endpoint with one API key. Switching is a one-line change to the model id, so you can route a fraction of traffic to each and compare bills directly.
Do MiniMax M3 and Qwen3.7 Plus support prompt caching?
Yes. Both bill cache reads below their input rate, so warm-prefix workloads cost less than the list rates suggest. The exact cache-read rows are in the pricing table above.