MiniMax M3 vs Qwen3.8 Max
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
minimax-m3 is the cheaper of the two on every line of the rate card, at $0.3 input and $1.2 output versus $2 and $6 for qwen3.8-max - roughly 6.7x and 5x less - and it also accepts video alongside text and image, carries a 1000000-token context with up to 524288 tokens of output, and lets you turn thinking off. qwen3.8-max costs more and caps output at 131072 tokens on a 983616-token context, so reach for it when you want Alibaba's newer August 2026 model with an explicit vision flag on text and image work. For high-volume, long-output or video-input jobs, minimax-m3 is the economical pick.
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 ByteDance DeepSeek DeepSeek Google Google Google Google MiniMax Moonshot Moonshot OpenAI OpenAI OpenAI OpenAI OpenAI OpenAI OpenAI Tencent Z.ai Z.ai Z.ai Z.ai
Pricing
| MiniMax M3 | Qwen3.8 Max | Δ | |
|---|---|---|---|
| Input / 1M tokens | $0.3 | $2 | 0.15× |
| Output / 1M tokens | $1.2 | $6 | 0.2× |
| Cache read / 1M tokens | $0.06 | $0.25 | 0.24× |
| Cache write | no separate charge | 1.25x | - |
Rates from the live catalogue at build time; each model page carries the current rate card.
Where they sit · input price per 1M tokens across all 91 chat models on this billing unit (log scale)
Capabilities
| MiniMax M3 | Qwen3.8 Max | |
|---|---|---|
| Tool calling | yes | yes |
| Thinking control | configurable | yes (vendor dial not published) |
| 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 | Qwen3.8 Max | |
|---|---|---|
| Input modalities | text image video | text image |
| Output modalities | text | text |
| Released | 2026-06-01 | 2026-08-03 |
| Context window | 1,000,000 | 983,616 |
| Max output | 524,288 | 131,072 |
| Thinking parameter |
| - |
| Accepted values | thinking.type
reasoning_split
| - |
| Default | adaptive: thinking on, with the model deciding when extra reasoning helps | - |
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 · Qwen3.8 Max
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
Qwen3.8 Max passed · 3 sentences
Enabling reasoning makes the model produce additional hidden steps before responding, and those tokens are billable. It also tends to lengthen each interaction because the model works through more possibilities before settling on an answer. Therefore, the bill doubled mainly due to higher compute and token usage per request, not necessarily because the number of requests doubled.
out 378 tok (+305 thinking) latency 8.6 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
Qwen3.8 Max passed · 8/8 cases
```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 best = max(best, cur) return best ``` The bug is that the original code starts `best` and `cur` at `0`, so it counts adjacent equal *transitions* rather than the number of items in the run. A run of length `n` has only `n - 1` equal-neighbor transitions, so single-element inputs return `0`, and runs that reach the end are undercounted by one. Initializing the current run to `1` for the first element, resetting it to `1` on a break, and updating `best` from that count fixes the off-by-one.
out 1616 tok (+1411 thinking) latency 34.7 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
Qwen3.8 Max passed · 5/5 fields, guidance null
{ "company": "Meridian Robotics", "quarter": "September quarter", "revenue_usd_m": 412, "growth_pct": 11, "guidance": null }
out 1199 tok (+1141 thinking) latency 24.4 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
Qwen3.8 Max passed · 120 words, 0 banned, 1 question
Today, our API gateway adds prompt caching across major model providers. It stores prompts and responses in one fast cache layer. Teams can lower token spend, reduce latency, and repeat reliable answers. The feature supports OpenAI, Anthropic, Google, and Mistral through one configuration. You can set retention rules, scope access, and invalidate entries quickly. How does your team maintain consistent results during provider outages? Approved cached responses keep applications stable while fallback routes recover. The dashboard shows hit rates, savings, latency, and provider usage. Engineers receive audit trails for every cached prompt, enabling safer testing. Product managers can compare cost trends before and after cache adoption. Start with a small route, then safely expand caching to production traffic right now.
out 2744 tok (+2591 thinking) latency 46.3 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="qwen3.8-max", # 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: "qwen3.8-max", // 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": "qwen3.8-max", # 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: "qwen3.8-max", // 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("qwen3.8-max") // 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 Qwen3.8 Max?
MiniMax M3 is cheaper on the "Input / 1M tokens" row ($0.3 vs $2, 6.7× apart). Other rows may point the other way; the table above carries the full rate card, and real cost depends on your mix.
Can I A/B test MiniMax M3 against Qwen3.8 Max 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 Qwen3.8 Max 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.