GMI: Models Intelligence, Performance & Price
This analysis is intended to support you in choosing the best model provided by GMI for your use-case.
Most Intelligent
MiMo-V2.6-Pro (BF16)
DeepSeek V4.1 Flash 1017 (max)
GLM-5.3 (max)
MiniMax-M3.1-FlashGLM-5.3-FlashIntelligence Index
Total 17 models
Fastest
DeepSeek V4 Flash Vision (max)DeepSeek V4.1 Flash 1017 (max)GLM-5.3 (max)DeepSeek V4.1 Flash (max)MiniMax-M3Output speed
Total 17 models
Lowest Price
MiMo-V2.6-Flash (BF16)GLM-5.3-FlashMiniMax-M3.1-Flash
Hy3DeepSeek V4.1 Flash (max)Blended price (per 1M tokens)
Total 17 models
GMI offers 17 models, each with different intelligence, performance, and pricing characteristics. Below is a comparison of the key metrics across models.
- For intelligence, the top models on GMI are MiMo-V2.6-Pro (BF16) (46), DeepSeek V4.1 Flash 1017 (max) (45), and GLM-5.3 (max) (45).
- For output speed, the fastest models are DeepSeek V4 Flash Vision (max) (173 t/s), DeepSeek V4.1 Flash 1017 (max) (153 t/s), and GLM-5.3 (max) (138 t/s).
- For latency, Qwen3.6 35B A3B (non-reasoning) FP8 (1.76s), Kimi K2.5 (non-reasoning) (3.17s), and DeepSeek V4 Flash Vision (max) (13.85s) offer the lowest time to first answer token.
- For pricing, MiMo-V2.6-Flash (BF16) ($0.06), GLM-5.3-Flash ($0.06), and MiniMax-M3.1-Flash ($0.10) offer the lowest blended prices per 1M tokens.
- For context window size, DeepSeek V4 Pro 0813 (max) (1M), MiniMax-M3 (1M), and DeepSeek V4.1 Flash 1017 (max) (1M) support the largest context windows on GMI.
Intelligence Evaluations
Artificial Analysis Intelligence Index
Evaluations
Agentic knowledge work, Elo
Knowledge
1 - hallucination rate
Agentic SaaS workflows
Agentic real-world work tasks, Elo
Reasoning & knowledge
Professional document reasoning, All-pass
Agentic coding & terminal use
Long context reasoning
Coding
Physics reasoning
Agentic scientific research workflows in a terminal
Agentic software engineering in open-source repositories
Quantitative analysis on spreadsheets & documents
Legal agentic work, Hallucination-Gated All-Pass Rate
Agentic business operations
Agentic interactive 3D experiences, Elo
Product and brand videos designed in code, Elo
Kubernetes incident root-cause analysis
Medical long context reasoning
Visual reasoning
Rich, varied, well-crafted language
Well-structured, scannable, answer-first
Truthful, acknowledges uncertainty, non-sycophantic
Use of filler and generic language
Intelligence Index vs. Price
Context Window
Context Window
Pricing
Intelligence Index vs. Price
Performance Summary
Output Speed vs. Price
Speed
Measured by Output Speed (tokens per second)
Output Speed
Latency
Measured by Time (seconds) to First Token
Latency: Time To First Answer Token
Cache Behaviour
Cache Hit Rate
Cost per Task vs. Cache Hit Rate
End-to-End Response Time
Seconds to output 500 tokens, calculated based on time to first token, 'thinking' time for reasoning models, and output speed
End-to-End Response Time vs. Price
Further Analysis | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
MiMo-V2.6-Pro (BF16) | 1M | Open | 46 | $0.12 | 78 | 2.71 | 34.78 | 25.65 | |||
DeepSeek V4.1 Flash 1017 (max) | 1.05M | Open | 45 | $0.23 | 153 | 2.16 | 18.52 | 13.09 | |||
GLM-5.3 (max) | 1M | Open | 45 | $1.42 | 138 | 2.18 | 20.30 | 14.49 | |||
MiniMax-M3.1-Flash | 1M | Proprietary | 43 | $0.19 | 111 | 1.48 | 24.06 | 18.06 | |||
GLM-5.3-Flash | 1M | Open | 42 | $0.24 | 60 | 2.87 | 44.37 | 33.20 | |||
DeepSeek V4.1 Flash (max) | 1M | Open | 39 | $0.18 | 135 | 1.95 | 20.46 | 14.81 | |||
MiMo-V2.6-Flash (BF16) | 524k | Open | 38 | $0.06 | 57 | 3.03 | 47.17 | 35.31 | |||
DeepSeek V4 Pro 0813 (max) | 1.05M | Open | 36 | $0.91 | 70 | 2.84 | 38.43 | 28.47 | |||
DeepSeek V4 Flash Vision (max) | 1M | Proprietary | 35 | $0.28 | 173 | 2.31 | 16.73 | 11.53 | |||
DeepSeek V4 Flash 0731 (max) | 1.05M | Open | 34 | $0.19 | 116 | 2.86 | 24.43 | 17.26 | |||
GLM-5.2 (max) (FP8) | 1.05M | Open | 34 | -- | -- | -- | -- | -- | |||
DeepSeek V4 Pro (max) | 1.05M | Open | 30 | -- | -- | -- | -- | -- | |||
MiniMax-M3 | 1.05M | Open | 29 | $0.36 | 117 | 1.30 | 22.61 | 17.05 | |||
 Kimi K2.6 FP8 | 262k | Open | 27 | -- | -- | -- | -- | -- | |||
GLM-5.1 (FP8) | 203k | Open | 26 | -- | -- | -- | -- | -- | |||
MiMo-V2.5-Pro | 1.05M | Open | 26 | -- | -- | -- | -- | -- | |||
Kimi K2.7 Code (FP8) | 65.5k | Open | 26 | $1.37 | 87 | 5.20 | 36.40 | 25.48 | |||
Hy3 | 262k | Open | 25 | $0.08 | 97 | 2.86 | 28.76 | 20.72 | |||
DeepSeek V4 Flash (high) | 1.05M | Open | 24 | -- | -- | -- | -- | -- | |||
DeepSeek V4 Flash (max) | 1.05M | Open | 24 | -- | -- | -- | -- | -- | |||
 Qwen3.5 27B (FP8) | 262k | Open | 23* | -- | -- | -- | -- | -- | |||
MiniMax-M2.5 FP8 | 197k | Open | 23* | -- | 58 | 2.00 | 45.12 | 34.50 | |||
MiniMax-M2.7 (FP8) | 197k | Open | 23 | -- | -- | -- | -- | -- | |||
Hy3-preview | 262k | Open | 23* | -- | -- | -- | -- | -- | |||
MiMo-V2.5 | 1.05M | Open | 22 | -- | -- | -- | -- | -- | |||
Kimi K2.5 (non-reasoning) | 262k | Open | 19* | -- | 48 | 3.17 | 13.62 | -- | |||
Qwen3.5 35B A3B (FP8) | 262k | Open | 19* | -- | -- | -- | -- | -- | |||
DeepSeek V4 Flash (non-reasoning) | 1.05M | Open | 19* | -- | -- | -- | -- | -- | |||
MiMo-V2.5-Pro (non-reasoning) | 1.05M | Open | 18* | -- | -- | -- | -- | -- | |||
Qwen3.6 35B A3B FP8 | 262k | Open | 18 | $0.31 | 110 | 1.82 | 55.33 | 48.98 | |||
Hy3-preview (non-reasoning) | 262k | Open | 17* | -- | -- | -- | -- | -- | |||
DeepSeek V3.2 (non-reasoning) | 164k | Open | 16* | -- | -- | -- | -- | -- | |||
Qwen3.6 35B A3B (non-reasoning) FP8 | 262k | Open | 15* | -- | 91 | 1.76 | 7.23 | -- | |||
Key definitions
Frequently Asked Questions
Common questions about GMI
GMI offers 17 models that we track: MiMo-V2.6-Pro (BF16), DeepSeek V4.1 Flash 1017 (max), GLM-5.3 (max), MiniMax-M3.1-Flash, GLM-5.3-Flash, DeepSeek V4.1 Flash (max), MiMo-V2.6-Flash (BF16), DeepSeek V4 Pro 0813 (max), DeepSeek V4 Flash Vision (max), DeepSeek V4 Flash 0731 (max), MiniMax-M3, Kimi K2.7 Code (FP8), Hy3, MiniMax-M2.5 FP8, Kimi K2.5 (non-reasoning), Qwen3.6 35B A3B FP8, and Qwen3.6 35B A3B (non-reasoning) FP8.
The most intelligent model available on GMI is MiMo-V2.6-Pro (BF16) with an Intelligence Index score of 46.
The fastest model on GMI by output speed is DeepSeek V4 Flash Vision (max) at 173.4 tokens per second.
The model with the lowest time to first answer token on GMI is Qwen3.6 35B A3B (non-reasoning) FP8 at 1.76s. Lower latency means faster initial response time.
The most affordable model on GMI by blended price is MiMo-V2.6-Flash (BF16) at $0.06 per 1M tokens (7:2:1 cache hit/input/output ratio).
Prices on GMI vary up to 14x across models, from $0.06 per 1M tokens for MiMo-V2.6-Flash (BF16) to $0.84 per 1M tokens for Kimi K2.5 (non-reasoning).
Yes, GMI offers an OpenAI-compatible API, making it easy to switch from OpenAI or use existing OpenAI SDK integrations.
Yes, all 17 models on GMI support JSON mode for structured output.
Yes, all 17 models on GMI support function calling (tool use).
Yes, GMI offers 15 reasoning models: MiMo-V2.6-Pro (BF16), DeepSeek V4.1 Flash 1017 (max), GLM-5.3 (max), MiniMax-M3.1-Flash, GLM-5.3-Flash, DeepSeek V4.1 Flash (max), MiMo-V2.6-Flash (BF16), DeepSeek V4 Pro 0813 (max), DeepSeek V4 Flash Vision (max), DeepSeek V4 Flash 0731 (max), MiniMax-M3, Kimi K2.7 Code (FP8), Hy3, MiniMax-M2.5 FP8, and Qwen3.6 35B A3B FP8. Reasoning models use extended thinking to work through complex problems before providing an answer.
Yes, 15 of 17 models on GMI are open weight models: MiMo-V2.6-Pro (BF16), DeepSeek V4.1 Flash 1017 (max), GLM-5.3 (max), GLM-5.3-Flash, DeepSeek V4.1 Flash (max), MiMo-V2.6-Flash (BF16), DeepSeek V4 Pro 0813 (max), DeepSeek V4 Flash 0731 (max), MiniMax-M3, Kimi K2.7 Code (FP8), Hy3, MiniMax-M2.5 FP8, Kimi K2.5 (non-reasoning), Qwen3.6 35B A3B FP8, and Qwen3.6 35B A3B (non-reasoning) FP8.
Yes, provider performance can vary over time due to infrastructure changes, load balancing, and updates. We continuously benchmark all providers and display historical performance trends in the "Over Time" charts.
When choosing a model on GMI, consider: intelligence (for quality-sensitive tasks), output speed (for throughput-intensive tasks), latency (for interactive applications requiring quick first responses), pricing (for cost-sensitive workloads), and features like context window size, JSON mode, or function calling support.