GitHub Copilot
Coding plans · reviewed by Stan K. · Source: data snapshot · data as of 2026-11-11 · scored at Pro
The plan everyone already has access to: $10 buys $15 of multi-model credits (Claude, GPT, Gemini, Grok, Kimi) with unlimited completions and a CLI — the credits just burn faster than GitHub admits.
#3 of 10 in AI Coding Plans
Tiers — what each dollar buys
| Tier | $/mo | Annual | Quota & notes |
|---|---|---|---|
| Free | $0 | — | 2,000 completions + 50 chats/mo; limited models |
| Pro scored | $10 | — | 1,500 credits ($15 value) · unlimited completions · Copilot CLI |
| Pro+ | $39 | — | 7,000 credits ($70) · Claude Code / Codex as third-party agents |
| Max | $100 | — | 20,000 credits ($200) |
Agent support: Copilot CLI + IDE agents — Copilot CLI ships on every paid tier; Pro+ additionally unlocks Claude Code and Codex as third-party agents. Still primarily GitHub's own surfaces (IDE extension, CLI, web).
PlanScore breakdown — why 68/100
10 plan criteria × 10 points; PlanScore = mean × 10. Low pillars are visible on purpose — that's the point of a score you can audit. Methodology →
What it does well
- ▸1,500 credits for $10 = ×1.5, rising to ×2 on Max — real (if modest) value across Claude, GPT, Gemini, Grok and Kimi
- ▸Unlimited code completions outside the credit system — the everyday feature never meters
- ▸No 5-hour or weekly walls: credits reset monthly, full stop
- ▸Copilot CLI on every paid tier; Claude Code and Codex unlocked as agents on Pro+
Where it costs you
- ▸Verified community range: the $15 credit bundle burns in ~2 days to 1 week of agent use
- ▸Annual plans retired (Apr 2026) — monthly only now
- ▸Premium-requests → credits transition (June 2026) is the second metering change in a year
- ▸Third-party agents gated behind Pro+ ($39)
Hidden traps — the stuff the pricing page doesn't say
Frontier-model agent requests cost real credits ($0.01 each, pricier models more): verified reports of the monthly bundle dying in 2 days. The unlimited part is completions — the agent features are exactly what's metered.
Premium requests (with per-model multipliers) were replaced by dollar credits in mid-2026; annual plans were retired in April 2026. The billing model has changed twice in one year.
Claude Code and Codex as powered agents only unlock at Pro+ ($39/mo) — the $10 tier's CLI is Copilot's own.
2,000 completions + 50 chats/month is an autocomplete teaser, not an agent plan.
Know the walls before you pay: read the traps above, then decide with eyes open.
See GitHub Copilot plansAffiliate link where available — never affects the score.
- official GitHub Copilot plans — credits, tiers (official docs) verified
- official Copilot pricing page (official) verified
- community HN: credit burn reports ×3 (via recon, 2026-11-11) verified
Data as of 2026-11-11. Prices and limits change monthly — report anything stale. How we verify: methodology.
Closest alternatives
The only plan that publishes its credit math — 12,000 credits/5h on Pro, half-price off-peak, works
Claude Pro / Max / Ultra — the model everyone else gets compared to, behind the most hated 5-hour +
Kimi Code from the $19 tier, a rolling 5-hour limit with NO weekly cap — and an October 2026 rebuild
GitHub Copilot plan review
In-depth analysis · last updated 2026-11-11
Tiers and credits
Free $0 (2,000 completions, 50 chats), Pro $10 = 1,500 credits ($15 value) (scored), Pro+ $39 = 7,000 ($70), Max $100 = 20,000 ($200). One credit = one cent of model usage across Claude (Haiku/Sonnet/Opus/Fable), GPT-5.3–6.1, Gemini 3.8–4, Grok 4.5–4.7 and Kimi K3. Annual plans were retired in April 2026.
The honest math
Pro multiplies your money ×1.5, Pro+ ×1.79, Max ×2.0 — the lowest multipliers among transparent plans here (OpenCode Go hits ×6, Z.ai ×4–8). What you're actually buying: unlimited completions (unmetered), the deepest IDE integration on the market, a monthly-only reset with no 5-hour and no weekly walls, and a CLI at every tier. If your work is IDE-assisted editing with occasional agent runs, the credit bundle is fine. If your work is agent-first, the verified community range is blunt: the bundle dies in two days to a week.
Worst value tier
Pro+ $39 for pure credits — $70 of metered usage is still agent-thin, and its real perk (Claude Code/Codex as agents) is about tool access, not volume.
Versus OpenCode Go and Z.ai
OpenCode Go ($10) pays ×6 in open CLIs across 32 models — Copilot Pro's ×1.5 only wins if the IDE integration and unlimited completions are worth the spread. Z.ai out-volumes it at every tier but carries the privacy record. Copilot is the safe default; it is not the value play.