text model comparison
MiMo-V2.6-Flash vs MiMo-V2.6-Pro
MiMo-V2.6-Flash and MiMo-V2.6-Pro are both current text models, from Xiaomi MiMo. MiMo-V2.6-Flash is 68% cheaper per 1M input tokens than MiMo-V2.6-Pro, $0.14 vs $0.435. On Arena Text, MiMo-V2.6-Flash scores 1448 Elo and MiMo-V2.6-Pro 1480 Elo (Arena, measured October 28, 2026). Both accept 1,000,000 tokens of context.
- Higher Arena Text
- MiMo-V2.6-Pro1480 Elo against 1448 Elo
- Lower price per 1M input tokens
- MiMo-V2.6-Flash$0.14 against $0.435
Data verified November 3, 2026 · Dataset 2026-11-06
Side by side
MiMo-V2.6-Flash and MiMo-V2.6-Pro, line by line
Prices in effect today, compared at the same tier wherever both list one. The better value in each row is highlighted.
| Fact | MiMo-V2.6-Flash | MiMo-V2.6-Pro |
|---|---|---|
| Price | ||
| Price per 1M input tokens | $0.14 (better) | $0.435 |
| Price per 1M output tokens | $0.28 (better) | $0.87 |
| Independent benchmarks | ||
| Arena Text | 1448 Elo (#35 of 52) | 1480 Elo (#19 of 52) (better) |
| Arena WebDev | 1637 Elo (#17 of 46) (better) | 1625 Elo (#20 of 46) |
| Specs | ||
| Provider | Xiaomi MiMo | Xiaomi MiMo |
| Release date | September 22, 2026 | September 22, 2026 |
| Access | API and open weights | API and open weights |
| Context window | 1,000,000 tokens | 1,000,000 tokens |
| Max output | 128,000 tokens | 128,000 tokens |
| Input | text, image, audio, video | text, image, audio, video |
| Output | text | text |
Scores:Arena · 2026-11-03
Provider results
What the labs report
Results tables a lab published with one of its models, where it lists both MiMo-V2.6-Flash and MiMo-V2.6-Pro. These are the lab's own figures, not independent measurements, and each table is read on its own: labs run benchmarks with different harnesses and budgets.
Thinking Machines Lab, Inkling 2 Small model card, November 5, 2026Thinking Machines Lab · 2026-11-06
| Benchmark | MiMo-V2.6-Flash | MiMo-V2.6-Pro |
|---|---|---|
| DeepSWE v1.1 | 67.9% | 71.9% (better) |
| Vibe Code Bench v1.1 | 79.0% | 85.2% (better) |
| MCP Atlaspublic set | 62.9% | 84.4% (better) |
| Finance Agent v2.1 | 63.3% | 64.3% (better) |
| Excel Modeling Benchmark v1.1 | 68.9% (better) | 67.4% |
| Tax Agent Bench | 59.9% | 64.9% (better) |
| Harvey LAB v1.1 | 13.3% | 13.8% (better) |
| Legal Research Bench | 38.0% | 47.1% (better) |
| CharXiv (RQ) | 66.2% | 74.2% (better) |
| Chartography | 7.5% | 10.8% (better) |
| SAGE v1.1 | 83.3% | 84.4% (better) |
| LongBench v2 | 60.7% | 68.0% (better) |
| ComplexConstraints v1.0 | 22.1% | 32.3% (better) |
| CyberBench v1.1 | 75.4% (better) | 72.9% |
Thinking Machines' table from the Inkling 2 Small model card, run at effort 0.99 (xhigh) with a 1M-token trajectory limit; where a benchmark's owner publishes scores (Datacurve, Scale AI, Surge, Mercor, Thoughtful, Vals), Thinking Machines uses them. DeepSeek-V4.1-Flash is the 1017 build. Left out here: the Artificial Analysis and Vals indexes, the rows Artificial Analysis ran (FrontierCode, Terminal-Bench, GDP.pdf, MMMU-Pro, HLE, CritPt, AA-LCR, AA-Omniscience, AutomationBench, EnterpriseOps-Gym, Tau 3 Banking), Arena and DesignArena ratings, PMPP-Hard (not a percentage), the safety rows, and JobBench, DAYJOB and Humanity's Sixth Sense, which we could not confirm are public.
Charts
Against every current text model
Where MiMo-V2.6-Flash and MiMo-V2.6-Pro land among the current models of their task, on every independent benchmark, on price and on context.
Where each stands
On Arena Text, MiMo-V2.6-Flash ranks #35 and MiMo-V2.6-Pro #19 of 52 current models. Each bar is the share of current models it beats: on score, price and context.
Independent benchmarks
Arena Text
MiMo-V2.6-Flash:#35of 52
1,448 Elo · ahead of 31%
MiMo-V2.6-Pro:#19of 52
1,480 Elo · ahead of 65%
Arena WebDev
MiMo-V2.6-Flash:#17of 46
1,637 Elo · ahead of 62%
MiMo-V2.6-Pro:#20of 46
1,625 Elo · ahead of 58%
Price and context
Price per 1M input tokens
MiMo-V2.6-Flash:#4of 58
$0.14 · cheaper than 95%
MiMo-V2.6-Pro:#17of 58
$0.435 · cheaper than 72%
Price per 1M output tokens
MiMo-V2.6-Flash:#1of 58
$0.28 · cheaper than 100%
MiMo-V2.6-Pro:#9of 58
$0.87 · cheaper than 86%
Context window
MiMo-V2.6-Flash:#27of 67
1M tokens · longer than 24%
MiMo-V2.6-Pro:#27of 67
1M tokens · longer than 24%
Quality against price
MiMo-V2.6-Flash scores 1,448 at $0.175; MiMo-V2.6-Pro scores 1,480 at $0.544 (blended price per 1M tokens). The line joins the current models no other beats on both price and Arena Text score; hover any point to name it.
Quality against price: the data
| Model | Provider | Blended price per 1M tokens | Arena Text (Elo) |
|---|---|---|---|
| Gemini 4 Carbon | $4 | 1,528 | |
| Gemini 4 Argon | $4 | 1,520 | |
| Claude Opus 5.5 | Anthropic | $8 | 1,509 |
| Muse Wonder | Meta | $8 | 1,507 |
| Claude Fable 5 | Anthropic | $20 | 1,504 |
| Claude Fable 5.1 | Anthropic | $20 | 1,504 |
| Gemini 3.8 Flash | $1.50 | 1,496 | |
| Muse Spark 1.3 | Meta | $2 | 1,495 |
| Claude Opus 5 | Anthropic | $10 | 1,492 |
| Kimi K3 | Moonshot Kimi | $6 | 1,488 |
| GPT-5.6 Sol | OpenAI | $8 | 1,485 |
| Gemini 3.6 Flash | $3 | 1,483 | |
| GPT-5.5 | OpenAI | $11 | 1,482 |
| Claude Opus 4.8 | Anthropic | $10 | 1,482 |
| Qwen3.8-Max | Alibaba Qwen | $3 | 1,482 |
| GPT-6.1 Sol | OpenAI | $4 | 1,481 |
| GLM-5.3 | Z.ai | $2.15 | 1,481 |
| MiMo-V2.6-Pro | Xiaomi MiMo | $0.54 | 1,480 |
| Claude Sonnet 5.5 | Anthropic | $4 | 1,477 |
| GLM-5.3-Flash | Z.ai | $0.24 | 1,477 |
| GPT-6 Astra | OpenAI | $20 | 1,476 |
| DeepSeek-V4.1-Flash | DeepSeek | $0.52 | 1,476 |
| GPT-5.4 | OpenAI | $5.63 | 1,475 |
| Gemini 3 Flash Preview | $1.13 | 1,473 | |
| Grok 4.20 | xAI | $1.56 | 1,472 |
| GPT-5.6 Terra | OpenAI | $4.50 | 1,467 |
| Grok 4.5 | xAI | $3 | 1,465 |
| DeepSeek-V4-Pro | DeepSeek | $1.98 | 1,465 |
| Claude Sonnet 5 | Anthropic | $4 | 1,462 |
| GPT-6 Sol | OpenAI | $4 | 1,458 |
| Gemini 3.5 Flash-Lite | $0.85 | 1,456 | |
| Grok 4.6 | xAI | $3 | 1,455 |
| GPT-5.6 Luna | OpenAI | $0.45 | 1,453 |
| MiMo-V2.6-Flash | Xiaomi MiMo | $0.18 | 1,448 |
| Step 5 Preview | StepFun | $1.43 | 1,448 |
| Grok 4.3 | xAI | $1.56 | 1,442 |
| GPT-6 Luna | OpenAI | $0.20 | 1,441 |
| Grok 4.7 | xAI | $3 | 1,441 |
| Mistral Large 4 | Mistral AI | $1.05 | 1,432 |
| Claude Haiku 5.5 | Anthropic | $0.20 | 1,425 |
| GPT-4.1 | OpenAI | $3.50 | 1,415 |
| Solar Pro 4 | Upstage | $0.52 | 1,388 |
| GPT-4.1 mini | OpenAI | $0.70 | 1,383 |
| GPT-4.1 nano | OpenAI | $0.18 | 1,322 |
Cost
What each costs
Example workloads at each provider's listed rates, then every other rate either one lists. The cheaper figure in each row is highlighted.
The same work, priced
For 10M input and 2M output tokens, MiMo-V2.6-Pro costs 3.1 times as much as MiMo-V2.6-Flash, at each provider's listed rates.
10M input and 2M output tokens
MiMo-V2.6-Pro costs 3.1× as much
- MiMo-V2.6-Flash
- $1.96
- MiMo-V2.6-Pro
- $6.09
The same, with 9M of the input tokens cached
MiMo-V2.6-Pro costs 3.0× as much
- MiMo-V2.6-Flash
- $0.73
- MiMo-V2.6-Pro
- $2.21
The same, through the batch API
MiMo-V2.6-Pro costs 3.1× as much
- MiMo-V2.6-Flash
- $0.98
- MiMo-V2.6-Pro
- $3.05
| Workload or rate | MiMo-V2.6-Flash | MiMo-V2.6-Pro |
|---|---|---|
| Example workloads | ||
| 10M input and 2M output tokens | $1.96 (better) | $6.09 |
| The same, with 9M of the input tokens cached | $0.73 (better) | $2.21 |
| The same, through the batch API | $0.98 (better) | $3.05 |
| Other listed rates | ||
| 1M input tokens, cached | $0.0028 (better) | $0.0036 |
| 1M input tokens, batch | $0.07 (better) | $0.2175 |
| 1M output tokens, batch | $0.14 (better) | $0.435 |
Your own bill will differ. These figures use list prices and example workloads. The AI Model Cost Audit reviews your real usage, shortlists models that score as well for each task for less, and estimates the savings.
Launch price $100, regular price $150
Verdict
When to choose each
MiMo-V2.6-Flash
Where it leads
- Lower input and output prices, $0.14 and $0.28 vs $0.435 and $0.87 per 1M tokens
- Higher Arena WebDev score, 1637 Elo vs 1625 Elo
- Cheaper cached input, $0.0028 vs $0.0036 per 1M tokens
- Cheaper batch processing, $0.98 vs $3.05 for 10M input and 2M output tokens
Best for
- High-frequency multimodal calls
- Low-cost agent workloads
Watch out for
- Xiaomi's own table puts it well behind the Pro model on exploitation tasks (ExploitBench 25.3 against 47.9)
- The API checkpoint (RL or MOPD weights) is not identified
MiMo-V2.6-Pro
Where it leads
- Higher Arena Text score, 1480 Elo vs 1448 Elo
Best for
- Long-horizon agent and coding work across modalities
- High-volume batch workloads
Watch out for
- The weights ship as two checkpoints (RL and MOPD), and Xiaomi does not say which one the API serves
- The mimo-v2.6-pro-ultraspeed id costs ten times as much and has no Batch API
From benchmark to production
Building on MiMo-V2.6-Flash or MiMo-V2.6-Pro?
Public scores rarely hold on your own data. We test both models on your real tasks, then build the agent or pipeline around the one that wins.
AI Model Cost AuditLaunch price $100, regular price $150What the audit includes
FAQ
MiMo-V2.6-Flash vs MiMo-V2.6-Pro FAQ
Which is cheaper, MiMo-V2.6-Flash or MiMo-V2.6-Pro?
MiMo-V2.6-Flash is 68% cheaper per 1M input tokens than MiMo-V2.6-Pro, $0.14 vs $0.435. For 10M input and 2M output tokens, MiMo-V2.6-Flash costs $1.96 and MiMo-V2.6-Pro $6.09 at standard rates.
Which is cheaper with prompt caching, MiMo-V2.6-Flash or MiMo-V2.6-Pro?
MiMo-V2.6-Flash charges $0.0028 per 1M cached input tokens, 98% below its standard input rate, and MiMo-V2.6-Pro $0.0036, 99% below. With 9M of 10M input tokens cached and 2M output tokens, MiMo-V2.6-Flash costs $0.73 and MiMo-V2.6-Pro $2.21.
Which scores higher on independent benchmarks, MiMo-V2.6-Flash or MiMo-V2.6-Pro?
MiMo-V2.6-Pro leads on Arena Text, 1480 Elo to 1448 Elo; MiMo-V2.6-Flash leads on Arena WebDev, 1637 Elo to 1625 Elo. Scores are independent evaluations, not either vendor's launch figures.
How do MiMo-V2.6-Flash and MiMo-V2.6-Pro compare in Thinking Machines Lab's own results?
In the Inkling 2 Small model card (November 5, 2026), MiMo-V2.6-Pro leads on DeepSWE v1.1, 71.9% to 67.9%; MiMo-V2.6-Pro leads on Vibe Code Bench v1.1, 85.2% to 79.0%; MiMo-V2.6-Pro leads on MCP Atlas (public set), 84.4% to 62.9%. These are Thinking Machines Lab's figures, not independent measurements.
Which has the larger context window, MiMo-V2.6-Flash or MiMo-V2.6-Pro?
Both accept 1,000,000 tokens of context. Both return up to 128,000 tokens per response.
Further reading
More on MiMo-V2.6-Flash and MiMo-V2.6-Pro
Related comparisons
- Claude Haiku 5.5 vs MiMo-V2.6-Flash
- Claude Sonnet 5.5 vs MiMo-V2.6-Pro
- GLM-5.3-Flash vs MiMo-V2.6-Flash
- DeepSeek-V4.1-Flash vs MiMo-V2.6-Pro
- GPT-6 Luna vs MiMo-V2.6-Flash
- Gemini 3 Flash Preview vs MiMo-V2.6-Pro
- Grok 4.3 vs MiMo-V2.6-Flash
- GLM-5.3-Flash vs MiMo-V2.6-Pro
- MiMo-V2.6-Flash vs Step 5 Preview
- GLM-5.3 vs MiMo-V2.6-Pro
- GPT-6.1 Terra vs MiMo-V2.6-Flash
- GPT-4.1 mini vs MiMo-V2.6-Pro
Production evaluation
Choosing between MiMo-V2.6-Flash and MiMo-V2.6-Pro?
Public benchmarks narrow the shortlist. We run both on your own tasks, data and cost limits, then build the routing and review controls to run the winner in production.
- A head-to-head evaluation on your real workload
- Routing between models by cost and difficulty
