Cerebras: Models Intelligence, Performance & Price
Analysis of Cerebras's models across key metrics including quality, price, output speed, latency, context window & more. This analysis is intended to support you in choosing the best model provided by Cerebras for your use-case.
Most Intelligent
Intelligence index
Total 4 models
Fastest
Output speed
Total 4 models
Lowest Price
Blended price (per 1M tokens)
Total 4 models
Cerebras offers 4 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 Cerebras are GLM-4.7 (34), GLM-4.7 (27), and gpt-oss-120b (high) (24).
- For output speed, the fastest models are gpt-oss-120b (high) (1,811 t/s), gpt-oss-120b (low) (1,751 t/s), and GLM-4.7 (846 t/s). Speed varies significantly across models, with a 115% difference between the fastest and slowest.
- For latency, GLM-4.7 (0.69s), gpt-oss-120b (high) (1.66s), and gpt-oss-120b (low) (1.70s) offer the lowest time to first answer token.
- For pricing, gpt-oss-120b (high) ($0.39), gpt-oss-120b (low) ($0.39), and GLM-4.7 ($2.30) offer the lowest blended prices per 1M tokens. Prices vary up to 5.9x across models.
- For context window size, gpt-oss-120b (high) (131k), gpt-oss-120b (low) (131k), and GLM-4.7 (131k) support the largest context windows on Cerebras.
- gpt-oss-120b (high) offers both the fastest output and best pricing, making it attractive for throughput-sensitive and cost-conscious applications. GLM-4.7 leads in intelligence for tasks that require the highest quality.
Highlights
Intelligence Evaluations
Artificial Analysis Intelligence Index
Intelligence Evaluations
Agentic real-world work tasks, (Elo-500)/2000
Agentic tool use
Agentic coding & terminal use
Coding
Reasoning & knowledge
Scientific reasoning
Physics reasoning
Knowledge
1 - hallucination rate
Long context reasoning
Agentic knowledge work, Elo
Agentic SaaS workflows
Legal agentic work, criterion pass rate
Agentic business operations
Instruction following
Long-horizon agentic tasks
Kubernetes incident root-cause analysis
Visual reasoning
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
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 | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
GLM-4.7 | 131k | Open | 34 | $1.19 | 841 | 0.74 | 3.71 | 2.38 | |||
 Gemma 4 31B | 131k | Open | 29 | $0.25 | -- | -- | -- | -- | |||
GLM-4.7 | 131k | Open | 27* | -- | 846 | 0.69 | 1.28 | -- | |||
gpt-oss-120b (high) | 131k | Open | 24 | $0.15 | 1,811 | 0.55 | 1.93 | 1.10 | |||
Gemma 4 31B | 131k | Open | 22 | $0.25 | -- | -- | -- | -- | |||
gpt-oss-120b (low) | 131k | Open | 15 | $0.05 | 1,751 | 0.56 | 1.99 | 1.14 | |||
Key definitions
Frequently Asked Questions
Common questions about Cerebras
Cerebras offers 4 models that we track: GLM-4.7, GLM-4.7, gpt-oss-120b (high), and gpt-oss-120b (low).
The most intelligent model available on Cerebras is GLM-4.7 with an Intelligence Index score of 34.
The fastest model on Cerebras by output speed is gpt-oss-120b (high) at 1,811.3 tokens per second.
The model with the lowest time to first answer token on Cerebras is GLM-4.7 at 0.69s. Lower latency means faster initial response time.
The most affordable model on Cerebras by blended price is gpt-oss-120b (high) at $0.39 per 1M tokens (7:2:1 cache hit/input/output ratio).
Prices on Cerebras vary up to 6x across models, from $0.39 per 1M tokens for gpt-oss-120b (high) to $2.30 per 1M tokens for GLM-4.7.
Yes, Cerebras offers an OpenAI-compatible API, making it easy to switch from OpenAI or use existing OpenAI SDK integrations.
Yes, all 4 models on Cerebras support JSON mode for structured output.
Yes, all 4 models on Cerebras support function calling (tool use).
Yes, Cerebras offers 3 reasoning models: GLM-4.7, gpt-oss-120b (high), and gpt-oss-120b (low). Reasoning models use extended thinking to work through complex problems before providing an answer.
Yes, all 4 models on Cerebras are open weight models.
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 Cerebras, 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.