Cerebras Code
Hacker ApprovedRanked 2 of 5 coding subscriptions by recommenders
Recommended for very high token allowances and fast inference on a flat monthly fee; the most common caveat is that the models are weaker than the newest options.
Recommended for very high token allowances and fast inference on a flat monthly fee; the most common caveat is that the models are weaker than the newest options.
40 people gave an opinion
- recommend it
- 28
- criticize it
- 12
- approval share
- 70%
Recommended in 10 of 13 months
- November 2025: recommended
- December 2025: recommended
- January 2026: recommended
- February 2026: recommended
- March 2026: recommended
- April 2026: recommended
- May 2026: recommended
- June 2026: recommended
- July 2026: recommended
- August 2026: not recommended
- September 2026: recommended
- October 2026: not recommended
- November 2026: not recommended (month in progress)
Owned for 3 months
Median of the 4 owners who said how long they have had it.
Each square is one year.
Why it carries the badge
- 28 people recommend itNeeds 10 or more, met.
- 2.3 recommenders for every criticNeeds 2 to 1 or better, met.
- Recommended in 10 of 13 monthsNeeds 4 or more, met.
The same three rules apply to every product, and the badge can’t be bought. How the badge works
What people say about it
The reasons people give, with the number of comments that give each one. Open a reason to read what they said.
Why people recommend it
Each mark is one comment
Very fast inference speed
“they are the fastest, and if you make heavy use their monthly subscription plan, then they're also the cheapest per token.”
Comment from 8 Nov 2025 in “You should write an agent” “if it takes care of the boring stuff, and it is fast while doing it, I am happy”
Comment from 8 Nov 2025 in “Cerebras Code now supports GLM 4.6 at 1000 tokens/sec” “the raw speed has considerably shifted the tipping point of time-savings for me”
Comment from 8 Nov 2025 in “Cerebras Code now supports GLM 4.6 at 1000 tokens/sec” - Comment from 16 Mar 2026 in “Ask HN: How is AI-assisted coding going for you professionally?”
- Comment from 8 Nov 2025 in “Cerebras Code now supports GLM 4.6 at 1000 tokens/sec”
- Comment from 8 Nov 2025 in “Cerebras Code now supports GLM 4.6 at 1000 tokens/sec”
- Comment from 8 Nov 2025 in “Cerebras Code now supports GLM 4.6 at 1000 tokens/sec”
- Comment from 8 Nov 2025 in “Cerebras Code now supports GLM 4.6 at 1000 tokens/sec”
Generous daily token limits
“Cerebras Code which gives you really generous token limits (like 24M per day on the 50 USD per month tier), though the GLM 4.7 model”
Comment from 7 Mar 2026 in “We Might All Be AI Engineers Now” “really generous token limits and amazing speed, but recently more downtime”
Comment from 16 Mar 2026 in “Ask HN: How is AI-assisted coding going for you professionally?” “I've also been moving around 100M tokens per week with Cerebras Code”
Comment from 18 Nov 2025 in “Gemini 3 Pro Model Card [pdf]” - Comment from 29 May 2026 in “Real-time LLM Inference on Standard GPUs: 3k tokens/s per request”
Low cost for heavy use
“they are the fastest, and if you make heavy use their monthly subscription plan, then they're also the cheapest per token.”
Comment from 8 Nov 2025 in “You should write an agent” “$600 per year is a trivial cost for a professional tool”
Comment from 8 Nov 2025 in “Cerebras Code now supports GLM 4.6 at 1000 tokens/sec” “The preserved concentration may be well worth the $50”
Comment from 8 Nov 2025 in “Cerebras Code now supports GLM 4.6 at 1000 tokens/sec” - Comment from 8 Nov 2025 in “Cerebras Code now supports GLM 4.6 at 1000 tokens/sec”
- Comment from 9 Nov 2025 in “Cerebras Code now supports GLM 4.6 at 1000 tokens/sec”
Good for simple edits
“I just use a mix of Cerebras Code for lots of fast/simpler edits and refactoring”
Comment from 26 Dec 2025 in “Nvidia buying AI chip startup Groq for about $20B in cash” “if it takes care of the boring stuff, and it is fast while doing it, I am happy”
Comment from 8 Nov 2025 in “Cerebras Code now supports GLM 4.6 at 1000 tokens/sec”
What people warn about
Each mark is one comment
Models seen as weaker
“I do have a Cerebras Code subscription (50 USD a month for a lot of tokens but only an okayish model)”
Comment from 9 Jan 2026 in “Anthropic bans use of API in OpenCode CLI tool” “It's GLM 4.7, GPT OSS 120B, or llama 3.1 8B so not exactly the latest or best models.”
Comment from 20 May 2026 in “How fast is N tokens per second really?” “50 USD a month for a lot of tokens but only an okayish model”
Comment from 9 Jan 2026 in “Anthropic bans use of API in OpenCode CLI tool” - Comment from 8 Nov 2025 in “Cerebras Code now supports GLM 4.6 at 1000 tokens/sec”
- Comment from 25 Oct 2026 in “Cerebras Deploys Trillion-Parameter AI Models”
- Comment from 8 Nov 2025 in “Cerebras Code now supports GLM 4.6 at 1000 tokens/sec”
Daily token limits run out fast
“They didn't seem to have prompt caching so I burnt through the daily 24M token limitations really quickly”
Comment from 20 May 2026 in “How fast is N tokens per second really?” “Still extremely fast, but in less than 1 hour I was able to use 7M input tokens”
Comment from 8 Nov 2025 in “Cerebras Code now supports GLM 4.6 at 1000 tokens/sec” Recent downtime and instability
“really generous token limits and amazing speed, but recently more downtime”
Comment from 16 Mar 2026 in “Ask HN: How is AI-assisted coding going for you professionally?” “their service wasn’t super stable last I tried”
Comment from 15 Apr 2026 in “Daily Claude outage is upon us. Waiting for Claude Status to update” Code edits sometimes break code
“Cerebras Code is nice for quick stuff and the sheer amount of tokens, but in KiloCode/... regularly messes up applying diff based edits”
Comment from 11 Dec 2025 in “GPT-5.2” “there were about 10 different places where the migration got fucked up and had to manually be fixed”
Comment from 17 Apr 2026 in “Claude Opus 4.7 costs 20–30% more per session” “where it fucked some code up and I had to go back to it and fix it twice”
Comment from 15 Feb 2026 in “Breaking the Spell of Vibe Coding”
Other coding subscriptions people recommend
The four most recommended other coding subscriptions, with the same counts for each, so you can compare like for like. The tick on each bar marks the 2 to 1 line the badge needs.
| Product | Recommend | Criticize | Approval | Months | Owned | Badge |
|---|---|---|---|---|---|---|
| Z.ai GLM Coding Subscription | 68 | 54 | 56% | 11 | 1 yr8 owners | — |
| Cerebras CodeThis page | 28 | 12 | 70% | 10 | 3 mo4 owners | Approved |
| MiniMax Coding Plan | 8 | 1 | 89% | 5 | — | — |
| Alibaba Coding Plan | 6 | 7 | 46% | 5 | — | — |
| Kimi coding plan | 5 | 1 | 83% | 4 | — | — |
What people said
40 of the 57 comments that mention it, newest first. Each links to the original comment.
“nowadays their qwen 3.8 gets roflstomped by Mercury 3 at both cost and performance”
“its fun but if the model is bad it just doesn’t do much useful”
“they seem to have stopped new signups”
“I used it with Cerebras inference at a time when it had a good coding plan at a low price”
“given how generous their token limits were”
“They didn't seem to have prompt caching so I burnt through the daily 24M token limitations really quickly”
“It's GLM 4.7, GPT OSS 120B, or llama 3.1 8B so not exactly the latest or best models.”
“It was really hard to keep up with how fast it works.”
“there were about 10 different places where the migration got fucked up and had to manually be fixed”
“their service wasn’t super stable last I tried”
“the future is something like Cerbaras Code that offers a speed up for single-threaded work”
“really generous token limits and amazing speed, but recently more downtime”
“Cerebras Code which gives you really generous token limits (like 24M per day on the 50 USD per month tier), though the GLM 4.7 model”
“where it fucked some code up and I had to go back to it and fix it twice”
“the "realtime" aspect of it does provide a qualitative difference”
“the other obvious answer is to plug in to the other claude code proxies”
“they have a subscription offering called Cerebras Code for $50/month”
“BUT: cached tokens count full, so if you have 100,000 tokens of context you can burn a minute of tokens in a few requests.”
“50 USD a month for a lot of tokens but only an okayish model”
“I do have a Cerebras Code subscription (50 USD a month for a lot of tokens but only an okayish model)”
“I just use a mix of Cerebras Code for lots of fast/simpler edits and refactoring”
“Even Cerebras Code that gives me a lot of tokens per day isn't enough for all tasks”
“Cerebras Code is nice for quick stuff and the sheer amount of tokens, but in KiloCode/... regularly messes up applying diff based edits”
“I've also been moving around 100M tokens per week with Cerebras Code”
“Cerebras is already much cheaper”
“$600 per year is a trivial cost for a professional tool”
“you see it happen almost real time”
“the $50 is actually pretty great value”
“they are the fastest, and if you make heavy use their monthly subscription plan, then they're also the cheapest per token.”
“It is definitely fun playing with these models at these speeds.”
“I find the fast models useless for deep coding work”
“does sped things up quite a bit”
“speed is great but quality wasn't great for me”
“the raw speed has considerably shifted the tipping point of time-savings for me”
“if it takes care of the boring stuff, and it is fast while doing it, I am happy”
“fast feedback of fast inference is a game changer”
“The preserved concentration may be well worth the $50”
“Still extremely fast, but in less than 1 hour I was able to use 7M input tokens”
“none of the Cerebras models at the $50 tier were worth anything for me”
“Code just flashes onto the page, it's pretty impressive.”
Counts come from 1,795,640 Hacker News comments, November 2025 to November 2026. How we count. Report a mistake.