coding on tomrochette.com

Cerebras Code

tom@tomrochette.com (Tom Rochette) — Sat, 26 Sep 2026 00:00:00 +0000

Cerebras Code is a subscription from chipmaker Cerebras that sells fast inference on one open coding model at $50/month (Pro) and $200/month (Max). The pitch is speed, and the record shows the speed claim and the quota fine print are the two things to verify before paying. Facts below verified as of 2026-09-26.

What it is
#

A hosted coding-inference plan on Cerebras wafer-scale hardware, consumed by pointing any OpenAI-compatible editor or agent (Cline, OpenCode, Crush, Cursor) at a Cerebras API key. It launched August 1, 2025 with Qwen3-Coder-480B advertised at up to 2,000 tokens per second and a 131k context window. As of 2026-09-26 the product page promotes GLM 4.7 at “1,000 tokens+ per second”, so the headline model has already been swapped once.

Status
#

Launched August 1, 2025 and drew 449 points and 172 comments on Hacker News the same day. Launch windows sold out repeatedly, and as of 2026-09-26 both Pro and Max are marked “sold out” on cerebras.ai/code, with a limited free trial still open. The model changed from Qwen3-Coder to GLM 4.7 between launch and now, which shows the plan follows whichever open model is fastest rather than committing to one family. I could not verify funding or subscriber counts.

Strengths
#

Speed is genuinely differentiated: even its harshest reviewer calls Cerebras the fastest provider of its model, bar none.
Flat monthly pricing with published daily allowances (24M tokens Pro, 120M Max) instead of per-token anxiety.
Bring-your-own-editor stance with OpenAI-compatible endpoints, no proprietary IDE lock-in.
By October 2025 the original tokens-per-minute caps had been raised in response to criticism.

Cautions
#

The marketing number did not survive contact: InfoWorld measured well under 500 tokens/second and often under 100, against the “up to 2,000” claim.
Undocumented throttles drove the experience: 300k TPM on Pro and 400k on Max produced 429 errors mid-session, and an early buyer reported a 7.5M-token daily cap hidden behind an advertised 1,000-request limit.
131k context is about half the model’s native window and demands careful context management.
At launch there was no prompt caching, which made agent loops expensive at the $2/1M API rate, and the usage console had defects (a Max purchase provisioning as Pro).
Cerebras declined InfoWorld’s request for comment on these issues.

Pricing
#

Pro costs $50/month with up to 24M tokens/day, and Max costs $200/month with up to 120M tokens/day. Both plans were marked sold out as of 2026-09-26; a free tier with limited tokens remains for connection testing. The underlying API price at launch was $2 per 1M input and $2 per 1M output on Qwen3-Coder. The current per-token table on cerebras.ai/pricing renders client-side and I could not extract it, so treat current API rates as unverified.

Price history
#

Date	Plan	Change	Source
2025-08-01	Pro / Max	Launched at $50/month (24M tokens/day) and $200/month (120M tokens/day)	Cerebras blog and HN thread
2025-09-15	Pro / Max	TPM caps (300k/400k) documented as the binding limit; prompt caching promised	InfoWorld review
2025-10-28	Pro / Max	Caps reported improved; Qwen3 deprecated in favor of GLM-4.6 from November	InfoWorld follow-up
2026-09-26	Pro / Max	Prices unchanged at $50/$200 but both marked sold out; model now GLM 4.7	cerebras.ai/code

Compared to
#

- Chutes is the cheap multi-model pay-as-you-go option; choose Chutes for price and variety, Cerebras when seconds per response matter.
- GLM Coding Plan undercuts Cerebras on quota per dollar and now serves the same GLM family; choose it for volume, Cerebras for raw tokens per second.

Bottom line
#

Recommended for developers whose bottleneck is iteration latency and who code in bursts that fit the daily token allowance. Not for heavy all-day agent runs, where the TPM throttles and 131k context cut the advertised advantage. Claim to disagree with: at the documented throttles, Max at $200 was worse value than four Pro accounts, because 4x300k TPM beats 400k TPM.

Changes
#

2026-09-26 - Created.

References
#

https://www.cerebras.ai/blog/introducing-cerebras-code - launch post: $50/$200 tiers, 24M/120M tokens/day, 2,000 tok/s claim, 131k context (fetched, HTTP 200).
https://www.cerebras.ai/blog/qwen3-coder-480b-is-live-on-cerebras - Qwen3-Coder launch, $2/1M API rate, “20x higher coding speed” claim (fetched, HTTP 200).
https://www.cerebras.ai/code - current product page: GLM 4.7 at 1,000+ tok/s, Pro and Max both marked “sold out” (fetched, HTTP 200, as of 2026-09-26).
https://news.ycombinator.com/item?id=44762959 - launch-day reception: 449 points, 172 comments, top comment flags missing caching (fetched, HTTP 200).
https://www.infoworld.com/article/4055909/ - critical review, Sep 15, 2025: disputed tok/s claims, TPM caps, 131k context, billing mixups, no vendor comment (fetched, HTTP 200).
https://www.infoworld.com/article/4075825/how-to-vibe-code-for-free-or-almost-free.html - follow-up: caps improved, Qwen3 deprecated for GLM-4.6, “fastest bar none” (fetched, HTTP 200).
https://www.reddit.com/r/LocalLLaMA/comments/1mfeazc/ - critical post, Aug 2, 2025: advertised 1,000 requests/day actually a 7.5M-token cap, “request” defined at ~8k tokens (retrieved, HTTP 200, via the Arctic Shift archive API).

GLM Coding Plan

tom@tomrochette.com (Tom Rochette) — Sat, 26 Sep 2026 00:00:00 +0000

The GLM Coding Plan is Z.ai’s subscription that sells access to the GLM model family inside coding agents, currently $18/$80/$168 per month for Lite/Pro/Max. It is the cheapest quota-per-dollar coding subscription I verified, but a multiplier system governs the quota, so the sticker price is only the start of the math. Facts below verified as of 2026-09-26.

What it is
#

A credit-based subscription from Z.ai (Zhipu) for GLM-5.3 and GLM-5.3-Flash, consumed inside supported coding tools (Claude Code, Cline, OpenCode, Kilo Code, Cursor, Z.ai’s own ZCode, and others). You point the tool at Z.ai’s Anthropic- or OpenAI-compatible endpoints with a plan key, and requests to older GLM versions are auto-routed to 5.3. Plans bundle exclusive MCP servers (vision, web search, web reader, Zread). Usage outside officially supported tools is contractually prohibited, which makes this a walled-garden subscription rather than an API credit pack.

Status
#

Actively developed and restructured twice in a year: prompt-based plans at launch in 2025, then a credits-based system for new subscribers on July 30, 2026, with legacy plans grandfathered to the end of their billing cycle. The plan rode the GLM release cadence (GLM-4.5 in 2025, GLM-5.2 in June 2026, GLM-5.3 in August 2026), and tier prices rose at each step. Promotional quota mechanics (off-peak discounts, Flash campaigns) changed repeatedly through 2026, so published value figures have a short shelf life. I found no reliable subscriber counts.

Strengths
#

Quota per dollar is unmatched in this category: Z.ai itself documents quota worth approximately 15-30x the monthly fee at API rates.
Lite at $18, or $12.60 effective on yearly billing, undercuts every $20 rival I checked.
Off-peak usage (outside Mon-Fri 14:00-18:00 UTC+8) deducts credits at 50%, a standing discount for most time zones.
Hard-stop quota means spend is fixed: when quota runs out, calls pause and your account balance is never touched.

Cautions
#

Subscriptions are non-refundable once purchased, per the official FAQ.
The flagship model burns quota up to 3x faster at peak (a usage multiplier, not a price change), so effective throughput can be a third of the headline prompts.
Strict supported-tools-only terms and special endpoints block custom integrations.
Third-party reviewers report slow and unreliable periods under load, and prompts route through China-based infrastructure, a blocker for residency-bound teams.

Pricing
#

Current tiers as of 2026-09-26: Lite $18/month (per Z.ai docs), Pro $80 and Max $168 per August 2026 snapshots of the subscribe page, with 20% off quarterly and 30% off yearly billing (effective $12.60/$56/$117.60 per month). The July 2026 credits system allocates 2,000/10,000 credits per 5 hours/week on Lite, 12,000/60,000 on Pro, 28,000/140,000 on Max. Credits deduct per token type with multipliers (GLM-5.3: input 6.9, cached input 1.7, output 24, over 10,000), plus per-call charges for bundled MCP tools. Under the legacy prompts system, Z.ai documented quota as roughly 80/400/1,600 prompts per 5-hour window, with one prompt estimated at 15-20 model invocations.

Price history
#

Date	Plan	Change	Source
2025-09	Lite / Pro	Baseline: launched at $6/$30 per month with first-month promos of $3/$15	Cline blog, Sep 18, 2025
2026-06-18	Lite / Pro / Max	List prices $18/$72/$160, yearly effective $12.60/$50.40/$112	HyScaler
2026-07-30	All plans	Credits-based plans replace prompt-based plans for new subscribers	docs.z.ai usage-revision notice
2026-08-14	Lite / Pro / Max	Pro to $80 and Max to $168; 20% quarterly and 30% yearly discounts	emergent.sh

Compared to
#

- Cerebras Code sells speed at $50/$200; choose it when latency dominates, GLM when volume per dollar dominates.
- Chutes sells cheap pay-as-you-go across many model families; choose it for flexibility, GLM for a large fixed quota on one frontier family.
Z.ai’s own ZCode is the intended surface, but any supported harness works with the same quota.

Bottom line
#

Recommended for high-volume, human-in-the-loop coding where quota per dollar is the binding constraint and you can work off-peak. Not for long-horizon autonomous agent runs (aggregator tracking puts GLM-5.2 at about half of Claude Opus 4.8’s SWE-Marathon score), Claude-ecosystem power users, or residency-bound teams. Claim to disagree with: the rational entry is a single month of $18 Lite and never the discounted yearly plan, because the plan is non-refundable and its terms changed twice in a year.

Changes
#

2026-09-26 - Created.

References
#

https://docs.z.ai/devpack/overview - current credit allowances (2,000/12,000/28,000 per 5h), credit formula, off-peak 50% rule, “starting at 18 USD” (fetched, HTTP 200).
https://docs.z.ai/devpack/notice/usage-revision - July 30, 2026 credits transition; legacy 15-30x value claim and per-plan prompt ceilings (fetched, HTTP 200).
https://docs.z.ai/devpack/faq - non-refundable policy, supported-tools restriction, no balance deduction, quota reset cards (fetched, HTTP 200).
https://cline.bot/blog/zai-cline-3-dollar-ai-coding - launch-era pricing: $6/$30 with $3/$15 first-month promos, 120/600 prompts per 5 hours (fetched, HTTP 200, Sep 18, 2025).
https://hyscaler.com/insights/glm-coding-plan-review/ - June 2026 tiers at $18/$72/$160, 3x peak multiplier, competitive framing (fetched, HTTP 200).
https://www.digitalapplied.com/blog/glm-coding-plan-worth-it-2026-value-analysis - critical analysis, Jul 3, 2026: multiplier math, non-refundability, China residency, long-horizon benchmark gap (fetched, HTTP 200).
https://emergent.sh/learn/glm-5-3-pricing - August 2026 tiers at $18/$80/$168 with yearly equivalents; GLM-5.3 has no per-token API rate yet (fetched, HTTP 200).

coding on tomrochette.com

Cerebras Code

What it is
#

Status
#

Strengths
#

Cautions
#

Pricing
#

Price history
#

Compared to
#

Bottom line
#

Changes
#

See also
#

References
#

GLM Coding Plan

What it is
#

Status
#

Strengths
#

Cautions
#

Pricing
#

Price history
#

Compared to
#

Bottom line
#

Changes
#

See also
#

References
#

coding on tomrochette.com

Cerebras Code

What it is #

Status #

Strengths #

Cautions #

Pricing #

Price history #

Compared to #

Bottom line #

Changes #

See also #

References #

GLM Coding Plan

What it is #

Status #

Strengths #

Cautions #

Pricing #

Price history #

Compared to #

Bottom line #

Changes #

See also #

References #

What it is
#

Status
#

Strengths
#

Cautions
#

Pricing
#

Price history
#

Compared to
#

Bottom line
#

Changes
#

See also
#

References
#

What it is
#

Status
#

Strengths
#

Cautions
#

Pricing
#

Price history
#

Compared to
#

Bottom line
#

Changes
#

See also
#

References
#