Choose Mistral: Codestral 2508 (batch) if…
- you need file / PDF input — to accept documents without your own parsing step; Qwen: Qwen3 30B A3B Instruct 2507 does not advertise it
Mistral: Codestral 2508 (batch) vs Qwen: Qwen3 30B A3B Instruct 2507
Neither wins outright: Mistral: Codestral 2508 (batch) is the pick when you need file / PDF input, Qwen: Qwen3 30B A3B Instruct 2507 when cost matters at scale.
Key numbers
| Spec | Mistral: Codestral 2508 (batch) | Qwen: Qwen3 30B A3B Instruct 2507 |
|---|---|---|
| Provider | Mistral AI | Qwen |
| Input price (per 1M tokens) | $0.15 | $0.10 |
| Output price (per 1M tokens) | $0.45 | $0.30 |
| Blended price (3:1 input/output) | $0.22 | $0.15 |
| Cheaper than … of all models we track | 68% | 73% |
| Context window | 256K tokens (~384 pages) | 262K tokens (~393 pages) |
| Larger context than … of all models | 45% | 48% |
| Max output per response | 205K tokens (~153,600 words) | 236K tokens (~176,900 words) |
| Input types | text, file | text |
| Output types | text | text |
| Free variant | No | No |
| Listed since | 2025-08-01 | 2025-07-29 |
Advertisement
What it costs for real workloads
Monthly cost at list price for five common workloads. Token counts per unit are stated so you can scale them to your own traffic. A workload that does not fit a model's context window is marked rather than priced.
| Workload (per month) | Mistral: Codestral 2508 (batch) | Qwen: Qwen3 30B A3B Instruct 2507 | Difference |
|---|---|---|---|
| Customer-support chatbot 50,000 conversations — each a 6-turn conversation (1,500 in / 400 out tokens) | $20.25 | $13.50 | Qwen: Qwen3 30B A3B Instruct 2507 saves $6.75 |
| RAG search over documents 20,000 questions — each a question answered from ~8 retrieved passages (6,000 in / 500 out tokens) | $22.50 | $15.00 | Qwen: Qwen3 30B A3B Instruct 2507 saves $7.50 |
| Coding / tool-using agent 2,000 tasks — each a multi-step task with tool calls and file context (60,000 in / 6,000 out tokens) | $23.40 | $15.60 | Qwen: Qwen3 30B A3B Instruct 2507 saves $7.80 |
| Batch summarisation 10,000 documents — each a 6,000-word report summarised to a page (8,000 in / 600 out tokens) | $14.70 | $9.80 | Qwen: Qwen3 30B A3B Instruct 2507 saves $4.90 |
| Long-document analysis 500 documents — each a 150-page contract or codebase read in one pass (100,000 in / 2,000 out tokens) | $7.95 | $5.30 | Qwen: Qwen3 30B A3B Instruct 2507 saves $2.65 |
Capabilities side by side
| Capability | Mistral: Codestral 2508 (batch) | Qwen: Qwen3 30B A3B Instruct 2507 | What it lets you do |
|---|---|---|---|
| Tool / function calling | ✓ Yes | ✓ Yes | call your APIs and run agent loops |
| Structured output (JSON schema) | ✓ Yes | ✓ Yes | return JSON that validates against your schema |
| Reasoning / thinking mode | — | — | spend extra tokens thinking before answering hard problems |
| Image input | — | — | read screenshots, charts and scanned pages |
| Audio input | — | — | take speech or audio directly |
| File / PDF input | ✓ Yes | — | accept documents without your own parsing step |
| Image output | — | — | generate images, not just text |
| Built-in web search | — | — | answer from live web results |
| Seed (reproducible sampling) | ✓ Yes | ✓ Yes | repeat a generation for tests and evals |
| Log-probabilities | — | ✓ Yes | read token confidence for classification and scoring |
From each model's published parameters and input/output types. “—” means not advertised, not proven absent.
Price history
- Mistral: Codestral 2508 (batch): input price cut 50% ($0.30 → $0.15/1M), output price cut 50% ($0.90 → $0.45/1M) on 2026-09-10.
- Qwen: Qwen3 30B A3B Instruct 2507: input price raised 52% ($0.05 → $0.10/1M), output price raised 36% ($0.19 → $0.30/1M) on 2026-10-09 (10 price changes since 2026-03-10).
| Date | Model | Input / 1M | Output / 1M |
|---|---|---|---|
| 2026-10-09 | Qwen: Qwen3 30B A3B Instruct 2507 | $0.10 | $0.30 |
| 2026-10-07 | Qwen: Qwen3 30B A3B Instruct 2507 | $0.05 | $0.19 |
| 2026-10-06 | Qwen: Qwen3 30B A3B Instruct 2507 | $0.10 | $0.30 |
| 2026-10-04 | Qwen: Qwen3 30B A3B Instruct 2507 | $0.05 | $0.19 |
| 2026-10-02 | Qwen: Qwen3 30B A3B Instruct 2507 | $0.10 | $0.30 |
| 2026-09-29 | Qwen: Qwen3 30B A3B Instruct 2507 | $0.05 | $0.19 |
| 2026-09-25 | Qwen: Qwen3 30B A3B Instruct 2507 | $0.10 | $0.30 |
| 2026-09-13 | Qwen: Qwen3 30B A3B Instruct 2507 | $0.05 | $0.19 |
Which should you choose?
Choose Qwen: Qwen3 30B A3B Instruct 2507 if…
- cost matters at scale — it is 33% cheaper on a typical 3:1 input/output mix ($13.50 vs $20.25 a month for 50,000 support chats)
- you need log-probabilities — to read token confidence for classification and scoring; Mistral: Codestral 2508 (batch) does not advertise it
Frequently asked questions
Is Mistral: Codestral 2508 (batch) better than Qwen: Qwen3 30B A3B Instruct 2507?
Neither wins outright: Mistral: Codestral 2508 (batch) is the pick when you need file / PDF input, Qwen: Qwen3 30B A3B Instruct 2507 when cost matters at scale. In short: Mistral: Codestral 2508 (batch) when you need file / PDF input; Qwen: Qwen3 30B A3B Instruct 2507 when cost matters at scale, you need log-probabilities.
Which is cheaper, Mistral: Codestral 2508 (batch) or Qwen: Qwen3 30B A3B Instruct 2507?
Qwen: Qwen3 30B A3B Instruct 2507 is cheaper: $0.15 vs $0.22 per 1M tokens at a typical 3:1 input/output mix, 33% less. For 50,000 support conversations a month that is $13.50 against $20.25.
How much do Mistral: Codestral 2508 (batch) and Qwen: Qwen3 30B A3B Instruct 2507 cost per million tokens?
Mistral: Codestral 2508 (batch) costs $0.15 per 1M input tokens and $0.45 per 1M output tokens; Qwen: Qwen3 30B A3B Instruct 2507 costs $0.10 input and $0.30 output. Output tokens are billed separately, so long answers shift the gap.
Which has the bigger context window, Mistral: Codestral 2508 (batch) or Qwen: Qwen3 30B A3B Instruct 2507?
Qwen: Qwen3 30B A3B Instruct 2507: 262K tokens, about 393 pages, against 256K (about 384 pages) for Mistral: Codestral 2508 (batch).
Do Mistral: Codestral 2508 (batch) and Qwen: Qwen3 30B A3B Instruct 2507 support function calling and JSON output?
Tool calling — Mistral: Codestral 2508 (batch): yes, Qwen: Qwen3 30B A3B Instruct 2507: yes. JSON-schema structured output — Mistral: Codestral 2508 (batch): yes, Qwen: Qwen3 30B A3B Instruct 2507: yes.
Can Mistral: Codestral 2508 (batch) or Qwen: Qwen3 30B A3B Instruct 2507 read images?
Neither accepts image input — both are text-in models.
Is Mistral: Codestral 2508 (batch) or Qwen: Qwen3 30B A3B Instruct 2507 free to use?
No free variant is listed for either; both are pay-per-token.
What is the maximum output length of Mistral: Codestral 2508 (batch) and Qwen: Qwen3 30B A3B Instruct 2507?
Mistral: Codestral 2508 (batch) can write up to 205K tokens (about 153,600 words) per response; Qwen: Qwen3 30B A3B Instruct 2507 up to 236K (about 176,900 words).
When were Mistral: Codestral 2508 (batch) and Qwen: Qwen3 30B A3B Instruct 2507 released?
Mistral: Codestral 2508 (batch) was listed on 2025-08-01; Qwen: Qwen3 30B A3B Instruct 2507 on 2025-07-29.
Has the price of Mistral: Codestral 2508 (batch) or Qwen: Qwen3 30B A3B Instruct 2507 changed?
Mistral: Codestral 2508 (batch): input price cut 50% ($0.30 → $0.15/1M), output price cut 50% ($0.90 → $0.45/1M) on 2026-09-10. Qwen: Qwen3 30B A3B Instruct 2507: input price raised 52% ($0.05 → $0.10/1M), output price raised 36% ($0.19 → $0.30/1M) on 2026-10-09 (10 price changes since 2026-03-10).
About the two models
Mistral: Codestral 2508 (batch)
Mistral's cutting-edge language model for coding released end of July 2025. Codestral specializes in low-latency, high-frequency tasks such as fill-in-the-middle (FIM), code correction and test generation. [Blog Post](https://mistral.ai/news/codestral-25-08)
Full Mistral: Codestral 2508 (batch) specs →Qwen: Qwen3 30B A3B Instruct 2507
Qwen3-30B-A3B-Instruct-2507 is a 30.5B-parameter mixture-of-experts language model from Qwen, with 3.3B active parameters per inference. It operates in non-thinking mode and is designed for high-quality instruction following, multilingual understanding, and...
Full Qwen: Qwen3 30B A3B Instruct 2507 specs →