mirror of
https://github.com/openai/codex.git
synced 2026-05-20 19:23:21 +00:00
f86d95a242
## Why The configurable `/statusline` and terminal title can display session token usage. That display was using the raw total token count, which includes cached input tokens, so it significantly overstated the token usage compared with the blended token count shown elsewhere (in `/status` and tracked in goals). This inconsistency resulted in user confusion. We don't want to report cached tokens because we don't charge for them and they are somewhat of an implementation detail that users shouldn't care about. ## What changed - Use `TokenUsage::blended_total()` for the `used-tokens` status surface item so cached input is excluded. - Add a brief comment to `tokens_in_context_window()` clarifying that it returns raw `total_tokens`, whose meaning depends on whether the caller has last-turn or accumulated usage.
90 lines
2.7 KiB
Rust
90 lines
2.7 KiB
Rust
//! TUI token usage models and display formatting.
|
|
|
|
use std::fmt;
|
|
|
|
use codex_protocol::num_format::format_with_separators;
|
|
use serde::Deserialize;
|
|
use serde::Serialize;
|
|
|
|
const BASELINE_TOKENS: i64 = 12000;
|
|
|
|
#[derive(Debug, Clone, Default, PartialEq, Eq, Serialize, Deserialize)]
|
|
pub struct TokenUsage {
|
|
pub input_tokens: i64,
|
|
pub cached_input_tokens: i64,
|
|
pub output_tokens: i64,
|
|
pub reasoning_output_tokens: i64,
|
|
pub total_tokens: i64,
|
|
}
|
|
|
|
impl TokenUsage {
|
|
pub fn is_zero(&self) -> bool {
|
|
self.total_tokens == 0
|
|
}
|
|
|
|
pub(crate) fn cached_input(&self) -> i64 {
|
|
self.cached_input_tokens.max(0)
|
|
}
|
|
|
|
pub(crate) fn non_cached_input(&self) -> i64 {
|
|
(self.input_tokens - self.cached_input()).max(0)
|
|
}
|
|
|
|
pub(crate) fn blended_total(&self) -> i64 {
|
|
(self.non_cached_input() + self.output_tokens.max(0)).max(0)
|
|
}
|
|
|
|
/// Returns the raw `total_tokens` value. For `last_token_usage`, this is the latest active
|
|
/// context size; for `total_token_usage`, this is the accumulated session total.
|
|
pub(crate) fn tokens_in_context_window(&self) -> i64 {
|
|
self.total_tokens
|
|
}
|
|
|
|
pub(crate) fn percent_of_context_window_remaining(&self, context_window: i64) -> i64 {
|
|
if context_window <= BASELINE_TOKENS {
|
|
return 0;
|
|
}
|
|
let effective_window = context_window - BASELINE_TOKENS;
|
|
let used = (self.tokens_in_context_window() - BASELINE_TOKENS).max(0);
|
|
let remaining = (effective_window - used).max(0);
|
|
((remaining as f64 / effective_window as f64) * 100.0)
|
|
.clamp(0.0, 100.0)
|
|
.round() as i64
|
|
}
|
|
}
|
|
|
|
#[derive(Debug, Clone, PartialEq, Eq, Serialize, Deserialize)]
|
|
pub(crate) struct TokenUsageInfo {
|
|
pub(crate) total_token_usage: TokenUsage,
|
|
pub(crate) last_token_usage: TokenUsage,
|
|
pub(crate) model_context_window: Option<i64>,
|
|
}
|
|
|
|
impl fmt::Display for TokenUsage {
|
|
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
|
|
write!(
|
|
f,
|
|
"Token usage: total={} input={}{} output={}{}",
|
|
format_with_separators(self.blended_total()),
|
|
format_with_separators(self.non_cached_input()),
|
|
if self.cached_input() > 0 {
|
|
format!(
|
|
" (+ {} cached)",
|
|
format_with_separators(self.cached_input())
|
|
)
|
|
} else {
|
|
String::new()
|
|
},
|
|
format_with_separators(self.output_tokens),
|
|
if self.reasoning_output_tokens > 0 {
|
|
format!(
|
|
" (reasoning {})",
|
|
format_with_separators(self.reasoning_output_tokens)
|
|
)
|
|
} else {
|
|
String::new()
|
|
}
|
|
)
|
|
}
|
|
}
|